benchmark stringlengths 2 20 | setting stringclasses 3
values | setting_name stringclasses 3
values | language stringclasses 3
values | suite stringclasses 7
values | suite_name stringclasses 7
values | motif stringclasses 16
values | motif_abbrev stringclasses 16
values | structural_complexity stringclasses 4
values | complexity_loc int64 9 535 β | complexity_functions int64 1 14 β | complexity_branches int64 0 26 β | complexity_loop_depth int64 1 74 β | complexity_score float64 0 9.35 β | in_paper_benchmark_set bool 2
classes | annotated_target_file stringlengths 4 24 β | annotated_start_line int64 1 1.19k β | annotated_end_line int64 23 1.47k β | annotated_regions_json stringlengths 56 123 β | baseline_code_path stringlengths 24 42 | baseline_code_bytes int64 2.11k 12.7M | baseline_code_sha256 stringlengths 64 64 | baseline_code_files int64 2 35 | baseline_kernel_files listlengths 0 3 | source_repo stringclasses 1
value | source_commit stringclasses 1
value | source_path stringlengths 6 24 | input_archive_path stringlengths 21 39 β | input_archive_bytes int64 518 3B β | input_archive_sha256 stringlengths 64 64 β | input_sizes listlengths 0 5 | n_input_sizes int64 0 5 | golden_output_sizes listlengths 0 5 | n_golden_output_sizes int64 0 5 | golden_output_bytes int64 0 21.8B | checker stringclasses 8
values | output_type stringclasses 2
values | output_dtype stringclasses 5
values | output_suffix stringclasses 7
values | atol float64 0 200 β | rtol float64 0 1 β | tolerance_source stringclasses 5
values | comparison_rule stringclasses 20
values | runtime_args_json stringlengths 15 19.6k β | upstream_suite stringclasses 7
values | upstream_url stringclasses 7
values | upstream_reference stringclasses 7
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2mm | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 28 | 1 | 0 | 4 | 1.038989 | true | 2mm.c | 77 | 107 | {"end_line": 107, "start_line": 77, "target_file": "2mm.c"} | baseline_code/EX1/2mm.zip | 9,599 | b0f247b828475da6d9207f63ee7fd7c0d3eb935f79db4585a1fce9559b8190c5 | 5 | [
"EX1_optimized_codes/2mm.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/2mm | input_data/EX1/2mm.zip | 22,352,766 | 089226938a4d59869412b63e458acad6168979b5d8cfa2f05baa192ef3be832d | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 48,563,067 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
3mm | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 35 | 1 | 0 | 5 | 1.574086 | true | 3mm.c | 73 | 112 | {"end_line": 112, "start_line": 73, "target_file": "3mm.c"} | baseline_code/EX1/3mm.zip | 9,194 | cbf4a76556a79975cd28eb31e50379dd32f48af7a2ae3623f8c016f0c8f04e8c | 4 | [
"EX1_optimized_codes/3mm.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/3mm | input_data/EX1/3mm.zip | 18,625,876 | 728d8d45f6bc360a8c32278eab8809b0be499bbab8d380b496d6d5c7ca0d6030 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 44,443,714 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
adi | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Stencil / Scientific Computing | SC | d1 | 56 | 1 | 0 | 2 | 0.482746 | true | adi.c | 68 | 130 | {"end_line": 130, "start_line": 68, "target_file": "adi.c"} | baseline_code/EX1/adi.zip | 9,759 | f6f38317cb87e3fbc17c2e588f9a26562a09405e077bb6ae94ce753f036a00bc | 4 | [
"EX1_optimized_codes/adi.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/adi | input_data/EX1/adi.zip | 241,174 | 9c925d48d2eb4ca3a1196da29a75967e5b563de8590d64c1654999f149789b01 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 25,472,560 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
aes | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Cryptography / Encoding | CE | d1 | 29 | 1 | 1 | 1 | 0.433599 | true | aes.c | 179 | 212 | {"end_line": 212, "start_line": 179, "target_file": "aes.c"} | baseline_code/EX1/aes.zip | 10,954 | 52f818e39efa05b704e0d91f6f52f1f70e9c7281dc1dd9d365514fb2c9af82c1 | 9 | [
"EX1_optimized_codes/aes.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/aes | input_data/EX1/aes.zip | 9,032 | cc4a0d85d8fe017a5c4c5ee5d3896055d4b4348396b79b1cadc6e7552c97fc3d | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 299 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
atax | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 21 | 1 | 0 | 3 | 0.503892 | true | atax.c | 66 | 88 | {"end_line": 88, "start_line": 66, "target_file": "atax.c"} | baseline_code/EX1/atax.zip | 9,072 | d39e1e30ade59be187aadb050651ea259f23792c04168ce652438ab155f10120 | 4 | [
"EX1_optimized_codes/atax.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/atax | input_data/EX1/atax.zip | 24,025 | ef97c2c93c10389685e3bde81f818ff519965be9ff843711c64a5fef92391695 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 43,850 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
atmux | EX1 | Serial CPU optimization | C | codee | Codee | Utility Benchmarks | UB | d1 | 9 | 1 | 0 | 3 | 0.449438 | true | atmux.c | 17 | 27 | {"end_line": 27, "start_line": 17, "target_file": "atmux.c"} | baseline_code/EX1/atmux.zip | 7,589 | d8a3ad025cde8e5424b45fbd60f9bb6da2964b91c24c37a6fd76123fc4ed089e | 8 | [
"EX1_optimized_codes/atmux.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/atmux | input_data/EX1/atmux.zip | 502,579 | 32aa7b2a300709ddc450fe9d61b46ad82c1a333ca9db44f1cd5badc3fc6795a5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 1,415,000 | codee | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["45000", "/dev/null"], "large": ["30000", "/dev/null"], "medium": ["20000", "/dev/null"], "mini": ["5000", "/dev/null"], "small": ["10000", "/dev/null"]} | Codee | https://www.codee.com/ | Codee open catalog of benchmarks. |
b+tree | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Graph Algorithms | GA | d3 | 110 | 2 | 11 | 5 | 6.085604 | true | null | null | null | {"end_line": [77, 101], "start_line": [1, 1], "target_file": ["kernel-cpu.c", "kernel-cpu-2.c"]} | baseline_code/EX1/b+tree.zip | 30,102 | a0ff2d4aa6db8cc00feaaf54d9cc72387e81bc6df0dcfeda0379f7493d72534e | 20 | [
"EX1_optimized_codes/kernel-cpu-2.c",
"EX1_optimized_codes/kernel-cpu.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/b+tree | input_data/EX1/b+tree.zip | 134,416,439 | c70de427d63bc282668b743a955a6147033c1ea14764b52152a384a235a195b9 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 19,386,282 | rodinia | text | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["file", "input_data/extra-large/input/mil.txt", "command", "input_data/extra-large/input/command.txt", "-o", "/dev/null"], "large": ["file", "input_data/large/input/mil.txt", "command", "input_data/large/input/command.txt", "-o", "/dev/null"], "medium": ["file", "input_data/medium/input/mil.txt", "comm... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
backprop | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Clustering / Machine Learning | CML | d1 | 29 | 2 | 0 | 2 | 0.652349 | true | backprop.c | 220 | 256 | {"end_line": 256, "start_line": 220, "target_file": "backprop.c"} | baseline_code/EX1/backprop.zip | 8,591 | 17fd76965c793e61270a41651056580166b6561aff0f5d1bea769e049be990e8 | 10 | [
"EX1_optimized_codes/backprop.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/backprop | input_data/EX1/backprop.zip | 711,501,009 | 07729392995524f798f432f7b33f80af85e976cf528b802af274d1a54e3ce891 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 788,792,160 | rodinia | binary | float32 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Binary file read as float32 (little-endian); shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["33554432", "-o", "/dev/null"], "large": ["8388608", "-o", "/dev/null"], "medium": ["2097152", "-o", "/dev/null"], "mini": ["65536", "-o", "/dev/null"], "small": ["1048576", "-o", "/dev/null"]} | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
basicmath-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Utility Benchmarks | UB | d1 | 34 | 1 | 2 | 1 | 0.776034 | true | cubic.c | 10 | 47 | {"end_line": 47, "start_line": 10, "target_file": "cubic.c"} | baseline_code/EX1/basicmath-mibench.zip | 12,477 | 77d1c4d08c3651374067abb03d6512fd35b50861fbbb535c8cbd08941c6ce617 | 14 | [
"EX1_optimized_codes/cubic.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/basicmath-mibench | input_data/EX1/basicmath-mibench.zip | 1,116,509,901 | b0c116b7248026d154d18333e477dbf7be343e782021294e445bce43fae5f1c8 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 5,023,864,851 | mibench | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["extra-large", "/dev/null"], "large": ["large", "/dev/null"], "medium": ["medium", "/dev/null"], "mini": ["mini", "/dev/null"], "small": ["small", "/dev/null"]} | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
bfs | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Graph Algorithms | GA | d3 | 140 | 3 | 13 | 3 | 6.616334 | true | bfs.cpp | 1 | 190 | {"end_line": 190, "start_line": 1, "target_file": "bfs.cpp"} | baseline_code/EX1/bfs.zip | 2,659 | 1dfe114b9d9730d7944ac1d99eaabdebea5614e24a5806d35385997dfa186429 | 3 | [
"EX1_optimized_codes/bfs.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/bfs | input_data/EX1/bfs.zip | 1,199,754,894 | e7f1d67a1115831aefc30d5356a89aa755ba32e346da16ac70c407729cbe9e1e | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 538,444,976 | rodinia | text | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["./input_data/extra-large/input/graph16M.txt", "-o", "/dev/null"], "large": ["./input_data/large/input/graph8M.txt", "-o", "/dev/null"], "medium": ["./input_data/medium/input/graph4M.txt", "-o", "/dev/null"], "mini": ["./input_data/mini/input/graph1M.txt", "-o", "/dev/null"], "small": ["./input_data/sm... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
bfs-bulk-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Graph Algorithms | GA | d1 | 40 | 3 | 3 | 2 | 2.130705 | true | bfs.c | 1 | 49 | {"end_line": 49, "start_line": 1, "target_file": "bfs.c"} | baseline_code/EX1/bfs-bulk-mach.zip | 10,723 | 44ea65ce4436e3205fac5b2a0195cf494063fd09f28e9e0b51c0ecf19a18fff4 | 8 | [
"EX1_optimized_codes/bfs.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/bfs-bulk-mach | input_data/EX1/bfs-bulk-mach.zip | 38,486 | 6f5b7a6c93a7675beeddebad26480aaac43830472c1871e3737f97b3dcfd6d57 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 132 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
bfs-queue-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Graph Algorithms | GA | d2 | 47 | 3 | 4 | 3 | 2.947052 | true | bfs.c | 1 | 56 | {"end_line": 56, "start_line": 1, "target_file": "bfs.c"} | baseline_code/EX1/bfs-queue-mach.zip | 8,215 | f3869a8fcff97a0760e2af3e0feaa5f142c4f2948bc4c184d38041475148581e | 7 | [
"EX1_optimized_codes/bfs.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/bfs-queue-mach | input_data/EX1/bfs-queue-mach.zip | 39,148 | 6ba2af7be30406602ac8ef8fad472468234b2caba665cefaf7d5ac37ebee3fc3 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 132 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
bicg | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 23 | 1 | 0 | 3 | 0.528366 | true | bicg.c | 74 | 98 | {"end_line": 98, "start_line": 74, "target_file": "bicg.c"} | baseline_code/EX1/bicg.zip | 6,710 | 3cf08a736541405e91add9bd4aa8c6a8942feb10721139f40891c0266a814cd2 | 4 | [
"EX1_optimized_codes/bicg.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/bicg | input_data/EX1/bicg.zip | 26,987 | d45b07a9204799ae741dcaaf902396fb4e9294c8618b1b08109821e72710fc0a | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 63,965 | polybench | text | float64 | .txt | 0.0005 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.0005, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
bilateral | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Image and Video Processing | IVP | d1 | 54 | 1 | 4 | 2 | 1.583272 | true | bilateralFilter-cpu.cpp | 95 | 160 | {"end_line": 160, "start_line": 95, "target_file": "bilateralFilter-cpu.cpp"} | baseline_code/EX1/bilateral.zip | 11,153 | be54c88dd0c497d7b0805d35a66c8f86c8fbf2828a0e4eca7d5e6b69ccec3be3 | 5 | [
"EX1_optimized_codes/bilateralFilter-cpu.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/bilateral | input_data/EX1/bilateral.zip | 4,163,837 | 90f6ab9bb59acb0179dc5af96c8fdc6387feb264c53412483c7163d294f4122b | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 6,144,000 | rodinia | binary | uint8 | .bin | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as uint8 (little-endian); shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["./input_data/nature_monte.bmp", "15", "10", "70", "/dev/null"], "large": ["./input_data/nature_monte.bmp", "10", "8", "50", "/dev/null"], "medium": ["./input_data/nature_monte.bmp", "5", "5", "30", "/dev/null"], "mini": ["./input_data/nature_monte.bmp", "1", "2", "3", "/dev/null"], "small": ["./input_... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
bitcount-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Utility Benchmarks | UB | d1 | 12 | 1 | 0 | 2 | 0 | true | bitcnts.c | 120 | 134 | {"end_line": 134, "start_line": 120, "target_file": "bitcnts.c"} | baseline_code/EX1/bitcount-mibench.zip | 12,165 | 8da2f820c5b0c380de9490152001d60949ca5c603d3ff1903dc8ff36f196b39b | 16 | [
"EX1_optimized_codes/bitcnts.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/bitcount-mibench | input_data/EX1/bitcount-mibench.zip | 3,562 | ca977583358c66f0e2a36018606e8f79f2c681b87c0af9e39cc2cf34a00f662c | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 378 | mibench | text | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"default": ["11250000", "--bits-only", "/dev/null"], "extra-large": ["912500000", "--bits-only", "/dev/null"], "large": ["312500000", "--bits-only", "/dev/null"], "medium": ["112500000", "--bits-only", "/dev/null"], "mini": ["11250000", "--bits-only", "/dev/null"], "small": ["11250000", "--bits-only", "/dev/null"]} | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
canneal-parsec | EX1 | Serial CPU optimization | C++ | parsec | PARSEC | Clustering / Machine Learning | CML | d1 | 38 | 1 | 3 | 2 | 1.106232 | true | annealer_thread.cpp | 19 | 64 | {"end_line": 64, "start_line": 19, "target_file": "annealer_thread.cpp"} | baseline_code/EX1/canneal-parsec.zip | 76,566 | 65b450db484fb08b91f769e30a3ed87768d9b9bb5b02fd04b9d198f1132ee265 | 35 | [
"EX1_optimized_codes/annealer_thread.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/canneal-parsec | input_data/EX1/canneal-parsec.zip | 256,138 | ce6c41e99b80bb76759646765fe2579102b921f12ea05bf91868a314ad0f6aee | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 2,544 | parsec | text | float64 | .txt | 0 | 0.000001 | checker constants (scripts/checkers/parsec.py) | Text summary; the candidate's 'Improvement:' value must be >= the reference value (absolute tolerance 1e-9, relative tolerance 1e-6). | {"extra-large": ["1", "1000000", "4000", "input_data/extra-large/input/input_extra-large.nets", "1000"], "large": ["1", "500000", "3500", "input_data/large/input/input_large.nets", "700"], "medium": ["1", "200000", "3000", "input_data/medium/input/input_medium.nets", "400"], "mini": ["1", "1000", "2000", "input_data/mi... | PARSEC | https://parsec.cs.princeton.edu/ | C. Bienia et al. The PARSEC Benchmark Suite: Characterization and Architectural Implications. PACT 2008. |
cfd | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Computational Fluid Dynamics | CFD | d2 | 195 | 1 | 4 | 3 | 3.758114 | true | euler3d-cpu.cpp | 194 | 403 | {"end_line": 403, "start_line": 194, "target_file": "euler3d-cpu.cpp"} | baseline_code/EX1/cfd.zip | 22,278 | 601fd6e1db431fb0fe288f27a991daa3bb879737cd8d4dc0b27c812d90fa91ae | 9 | [
"EX1_optimized_codes/euler3d-cpu.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/cfd | input_data/EX1/cfd.zip | 56,082,301 | 9fe7d1ca9fc031108d2c0c6403f48e0bc023c5d619234930b1158a97bf332ca9 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 14,653,751 | rodinia | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["input_data/extra-large/input/missile.domn.0.2M", "/dev/null"], "large": ["input_data/large/input/fvcorr.domn.193K", "/dev/null"], "medium": ["input_data/medium/input/fvcorr.domn.097K", "/dev/null"], "mini": ["input_data/mini/input/avg/nasa01600.domn.0.2M", "/dev/null"], "small": ["input_data/small/inp... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
cholesky | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 20 | 1 | 0 | 2 | 0.042217 | true | cholesky.c | 84 | 108 | {"end_line": 108, "start_line": 84, "target_file": "cholesky.c"} | baseline_code/EX1/cholesky.zip | 5,201 | fe70619ca642ae72b34cb8216681fe49c334f69e818b8d620a9c1a317b7e1151 | 4 | [
"EX1_optimized_codes/cholesky.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/cholesky | input_data/EX1/cholesky.zip | 1,597,610 | f0fc580cfe431f898c9338a3b252ed3ef208b1f7a1690cfdeb08cbe434ea49c3 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 50,964,440 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
correlation | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Statistical Computations | STC | d1 | 47 | 1 | 1 | 3 | 1.103302 | true | correlation.c | 67 | 126 | {"end_line": 126, "start_line": 67, "target_file": "correlation.c"} | baseline_code/EX1/correlation.zip | 8,125 | 3418320464f91a69da271b0035160b79ac9228f989584c80758a99e35969337e | 4 | [
"EX1_optimized_codes/correlation.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/correlation | input_data/EX1/correlation.zip | 385,950 | 5899fbc8f0c89893ebde08584ed4328a0ef5b8467539a818de059362a24b8db4 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 41,737,550 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
coulomb | EX1 | Serial CPU optimization | C | codee | Codee | N-body Methods | NBM | d1 | 24 | 1 | 0 | 2 | 0.091165 | true | coulomb.c | 16 | 40 | {"end_line": 40, "start_line": 16, "target_file": "coulomb.c"} | baseline_code/EX1/coulomb.zip | 6,266 | 02392a215e608b5229d4d3ecc1166ccc93b3566becc160b28570ff3f5d8971e3 | 6 | [
"EX1_optimized_codes/coulomb.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/coulomb | input_data/EX1/coulomb.zip | 1,945,659 | fbce127a081230e054b5ddfd6b28978da190fa07046868f5706c595b668ea7bb | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 4,751,500 | codee | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["400", "1", "0.1", "/dev/null"], "large": ["350", "1", "0.1", "/dev/null"], "medium": ["300", "1", "0.1", "/dev/null"], "mini": ["200", "1", "0.1", "/dev/null"], "small": ["250", "1", "0.1", "/dev/null"]} | Codee | https://www.codee.com/ | Codee open catalog of benchmarks. |
covariance | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Statistical Computations | STC | d1 | 30 | 1 | 0 | 5 | 1.512901 | true | covariance.c | 65 | 98 | {"end_line": 98, "start_line": 65, "target_file": "covariance.c"} | baseline_code/EX1/covariance.zip | 7,816 | 18e70e029bca28acc9f9bfbb0a3ffe5e8757573440cb70a116ace214cfea00e0 | 4 | [
"EX1_optimized_codes/covariance.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/covariance | input_data/EX1/covariance.zip | 38,274,051 | e19c52580d35aa2b93a8077cd9037845c311afb452e9f56fe54c0b2c84a87d12 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 77,881,278 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
cutcp | EX1 | Serial CPU optimization | C | parboil | Parboil | N-body Methods | NBM | d4 | 172 | 1 | 15 | 8 | 8.120977 | true | cutcpu.c | 1 | 221 | {"end_line": 221, "start_line": 1, "target_file": "cutcpu.c"} | baseline_code/EX1/cutcp.zip | 19,261 | d6d1dfc47f7f965043946c77af881491341d59128b87cf28b847e9a8fedb8309 | 13 | [
"EX1_optimized_codes/cutcpu.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/cutcp | input_data/EX1/cutcp.zip | 8,718,403 | 9d4ba79274d9a238244a714219ee1c650188a2ee6c16ff3f8547b91917c62e99 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 1,897,160 | parboil | binary | float32 | .dat | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as float32 (little-endian); shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"default": ["-i", "input_data/mini/input/input_0000.pqr", "-o", "/dev/null"], "extra-large": ["-i", "input_data/extra-large/input/input_0000.pqr", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/watbox.sl100.pqr", "-o", "/dev/null"], "medium": ["-i", "input_data/medium/input/input_0000.pqr", "-o", "/dev/nu... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
deriche | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Image and Video Processing | IVP | d1 | 80 | 1 | 0 | 2 | 0.776432 | true | deriche.c | 68 | 157 | {"end_line": 157, "start_line": 68, "target_file": "deriche.c"} | baseline_code/EX1/deriche.zip | 8,338 | 48de51363036ba066415e5214bf908cdac0507a8541e5e6321235e3bb7fbe6ce | 4 | [
"EX1_optimized_codes/deriche.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/deriche | input_data/EX1/deriche.zip | 20,360,462 | b2b660e35e1fd9513181523bb38ac7a6c538798710d7e7e5bcaffd22a1f17828 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 216,946,931 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
dijkstra-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Graph Algorithms | GA | d1 | 45 | 1 | 3 | 4 | 2.090766 | true | dijkstra_parameterized.c | 103 | 153 | {"end_line": 153, "start_line": 103, "target_file": "dijkstra_parameterized.c"} | baseline_code/EX1/dijkstra-mibench.zip | 10,907 | e879d41906dee006d606f66aa50e5e6653d12e5ec6a9f031b689c08ded5b9026 | 9 | [
"EX1_optimized_codes/dijkstra_parameterized.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/dijkstra-mibench | input_data/EX1/dijkstra-mibench.zip | 7,374,780 | 32ad0af54584fc1efc41fcd2e8e6ebc8eb6ae68db8b74ace1a29f019f127cb4b | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 30,153 | mibench | text | int32 | .dat | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["./input_data/extra-large/input/input.dat"], "large": ["./input_data/large/input/input.dat"], "medium": ["./input_data/medium/input/input.dat"], "mini": ["./input_data/mini/input/input.dat"], "small": ["./input_data/small/input/input.dat"]} | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
doitgen | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Statistical Computations | STC | d1 | 19 | 1 | 0 | 3 | 0.479419 | true | doitgen.c | 67 | 87 | {"end_line": 87, "start_line": 67, "target_file": "doitgen.c"} | baseline_code/EX1/doitgen.zip | 7,876 | 496fc5ac20435ea9cd43ccb5fd6b17069f5c050dc006ef30b5b6b6269045c209 | 4 | [
"EX1_optimized_codes/doitgen.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/doitgen | input_data/EX1/doitgen.zip | 35,565,520 | 9a108a48d785eff642d870e13b395c0ff25e7ff10329a2065dc10a21baef896f | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 110,894,591 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
durbin | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Statistical Computations | STC | d1 | 31 | 1 | 0 | 2 | 0.176823 | true | durbin.c | 62 | 97 | {"end_line": 97, "start_line": 62, "target_file": "durbin.c"} | baseline_code/EX1/durbin.zip | 4,556 | f851933a2a08e2f608f315c4fb7ef97f669ed780a727f988c3efc7da77fba07a | 4 | [
"EX1_optimized_codes/durbin.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/durbin | input_data/EX1/durbin.zip | 6,885 | 3d88a49549740a059d6eb742f67b9d0627df1bfddff0ae8aa900c0fe918a88d0 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 36,811 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
fdtd-2d | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Stencil / Scientific Computing | SC | d1 | 28 | 1 | 0 | 7 | 2.140113 | true | fdtd-2d.c | 91 | 121 | {"end_line": 121, "start_line": 91, "target_file": "fdtd-2d.c"} | baseline_code/EX1/fdtd-2d.zip | 4,888 | 122776f898b1dc4985490484a5bff705b5c4ff62559227bd321825062dc70564 | 4 | [
"EX1_optimized_codes/fdtd-2d.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/fdtd-2d | input_data/EX1/fdtd-2d.zip | 55,766,480 | 217aff8e860683b5465aac3bfee6bbc0b04f02a6fb9c466cc52010734a1553d9 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 135,314,665 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
fft-strided-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Spectral Methods | SM | d1 | 35 | 3 | 1 | 2 | 1.507021 | true | fft.c | 1 | 44 | {"end_line": 44, "start_line": 1, "target_file": "fft.c"} | baseline_code/EX1/fft-strided-mach.zip | 6,027 | 695cbf053eaad0230777da8762b40548266a28e54847f4aff39cd75d33618025 | 7 | [
"EX1_optimized_codes/fft.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/fft-strided-mach | input_data/EX1/fft-strided-mach.zip | 242,819 | 34edc59661fc3d7d6042bb5572b74ea42d798ac35a2a54e000a78216fc056c03 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 202,610 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
fft-transpose-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Spectral Methods | SM | d2 | 250 | 1 | 0 | 1 | 2.5 | true | fft.c | 114 | 415 | {"end_line": 415, "start_line": 114, "target_file": "fft.c"} | baseline_code/EX1/fft-transpose-mach.zip | 11,133 | 74c642a349effe452de559196bef8efd8567c62cf4646e5fa31d95d954d49313 | 8 | [
"EX1_optimized_codes/fft.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/fft-transpose-mach | input_data/EX1/fft-transpose-mach.zip | 107,771 | 8b3176571fd5d1ad8a1e078e50ef069a8cffd852b81ecde6f3e656d4bf7feb3d | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 100,590 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
floyd-warshall | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dynamic Programming | DP | d1 | 15 | 1 | 1 | 3 | 0.730688 | true | floyd-warshall.c | 53 | 69 | {"end_line": 69, "start_line": 53, "target_file": "floyd-warshall.c"} | baseline_code/EX1/floyd-warshall.zip | 3,652 | d8308d403edddca72b78ba5fc2b73f7ec312a046eb5ceccbf99b25aa1dd22750 | 3 | [
"EX1_optimized_codes/floyd-warshall.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/floyd-warshall | input_data/EX1/floyd-warshall.zip | 9,456,323 | be5ad8a92e4658609b29d7bf386fd302cdefa6278f791825f11b0b75a3012836 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 80,946,690 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
gemm | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 20 | 1 | 0 | 3 | 0.491656 | true | gemm.c | 73 | 93 | {"end_line": 93, "start_line": 73, "target_file": "gemm.c"} | baseline_code/EX1/gemm.zip | 4,613 | b14d84a4db438f2d5f14310d5b9cc504f6a27ad48957e2adcf58db2493ec23af | 4 | [
"EX1_optimized_codes/gemm.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/gemm | input_data/EX1/gemm.zip | 16,224,600 | de67a0c7ca64a585b268cccfb183df916d619dd1b99d7cd2995f7d9a730fd7a6 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 40,496,483 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
gemm-blocked-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Dense Linear Algebra | DLA | d1 | 30 | 3 | 0 | 2 | 1.164586 | true | gemm.c | 1 | 36 | {"end_line": 36, "start_line": 1, "target_file": "gemm.c"} | baseline_code/EX1/gemm-blocked-mach.zip | 7,872 | ebe403bbc8c52fc1d9f967e306b3d9643a206468375bbf3988a8e8f6de3d758f | 6 | [
"EX1_optimized_codes/gemm.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/gemm-blocked-mach | input_data/EX1/gemm-blocked-mach.zip | 591,691 | 6c3e4f928964b0f1249fe2b6d148d17aa45b900054d9ed01c34f90170670e9c2 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 409,615 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
gemm-ncubed-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Dense Linear Algebra | DLA | d1 | 27 | 3 | 0 | 2 | 1.127876 | true | gemm.c | 1 | 33 | {"end_line": 33, "start_line": 1, "target_file": "gemm.c"} | baseline_code/EX1/gemm-ncubed-mach.zip | 7,968 | 353ce4f333ee23ded936e4a73b16669767538a1883dc2c04485e4d5ff26cffee | 6 | [
"EX1_optimized_codes/gemm.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/gemm-ncubed-mach | input_data/EX1/gemm-ncubed-mach.zip | 591,996 | a7a707b4196275707bc01453a258ff6b3d106091acf3f3c932ced2a40eba42d7 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 409,615 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
gemver | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 29 | 1 | 0 | 7 | 2.152349 | true | gemver.c | 85 | 119 | {"end_line": 119, "start_line": 85, "target_file": "gemver.c"} | baseline_code/EX1/gemver.zip | 4,768 | e5a4ff843064eac2357539e7a419421a6762d97c79ff5a3c337144357135d250 | 4 | [
"EX1_optimized_codes/gemver.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/gemver | input_data/EX1/gemver.zip | 53,415 | 26c32d361a7a936e7d67d20c845006791c91ee9b3f2dfcbc0f4115364c48c089 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 98,870 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
gesummv | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 25 | 1 | 0 | 2 | 0.103402 | true | gesummv.c | 72 | 98 | {"end_line": 98, "start_line": 72, "target_file": "gesummv.c"} | baseline_code/EX1/gesummv.zip | 4,548 | d5e6483bb4f6191676421e1e818fdf44ae876cde803d909ac6f8f6a84acd84ad | 4 | [
"EX1_optimized_codes/gesummv.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/gesummv | input_data/EX1/gesummv.zip | 16,699 | 1f0a121f289aac67599861603e4ee3f6ac97d44eb57a127b4275c89c733436f9 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 34,546 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
gramschmidt | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 28 | 1 | 0 | 4 | 1.038989 | true | gramschmidt.c | 80 | 110 | {"end_line": 110, "start_line": 80, "target_file": "gramschmidt.c"} | baseline_code/EX1/gramschmidt.zip | 4,696 | 1a4005f077a04f697f6040b3d037a8c936b3a740da5ce45fb8a38930f107ad09 | 4 | [
"EX1_optimized_codes/gramschmidt.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/gramschmidt | input_data/EX1/gramschmidt.zip | 13,157,181 | 9df037c95c22ccb061b8aec4c77f6d09881f114b47ccc6273f515f8794d704ca | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 79,581,363 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
haccmk | EX1 | Serial CPU optimization | C | codee | Codee | N-body Methods | NBM | d1 | 34 | 3 | 2 | 2 | 1.776034 | true | Step10_orig.c | 1 | 48 | {"end_line": 48, "start_line": 1, "target_file": "Step10_orig.c"} | baseline_code/EX1/haccmk.zip | 9,296 | e7a54f14b270df58a512fc649495512802b11ebf823276d5c625a7d203a4917e | 14 | [
"EX1_optimized_codes/Step10_orig.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/haccmk | input_data/EX1/haccmk.zip | 518 | 24d61cf2428cf52dddc78403d4beeab6c8c5c151811e00774aa128588bb57bb9 | [] | 0 | [] | 0 | 0 | codee | text | float64 | .txt | 200 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=200, rtol=0). | {"default": []} | Codee | https://www.codee.com/ | Codee open catalog of benchmarks. |
heartwall | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Medical / Scientific Computing | MSC | d4 | 414 | 3 | 16 | 4 | 8.898876 | true | kernel.c | 1 | 596 | {"end_line": 596, "start_line": 1, "target_file": "kernel.c"} | baseline_code/EX1/heartwall.zip | 31,094 | 69c47fe806b06566d7f87965c7b286df61c16699a2ceade0f7636f1e7993af7b | 11 | [
"EX1_optimized_codes/kernel.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/heartwall | input_data/EX1/heartwall.zip | 10,684,885 | 1d15c88907927187e1178a2f6f9de1f8814613b1970471e93898983e85262f3e | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 66,900 | rodinia | text | float64 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"extra-large": ["input_data/test.avi.part00", "50", "/dev/null"], "large": ["input_data/test.avi.part00", "40", "/dev/null"], "medium": ["input_data/test.avi.part00", "30", "/dev/null"], "mini": ["input_data/test.avi.part00", "10", "/dev/null"], "small": ["input_data/test.avi.part00", "20", "/dev/null"]} | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
heat-3d | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Stencil / Scientific Computing | SC | d1 | 32 | 1 | 0 | 2 | 0.18906 | true | heat-3d.c | 65 | 98 | {"end_line": 98, "start_line": 65, "target_file": "heat-3d.c"} | baseline_code/EX1/heat-3d.zip | 4,722 | 3572c292253f2112ba0c20d25d918aecf754011c47760dd6c31378934fd39f67 | 4 | [
"EX1_optimized_codes/heat-3d.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/heat-3d | input_data/EX1/heat-3d.zip | 2,821,881 | b51fe9c3394fd0dbf641eb887f122c3ced4cbeea78b158f76b705418f79529a3 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 57,662,975 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
histo | EX1 | Serial CPU optimization | C | parboil | Parboil | Utility Benchmarks | UB | d2 | 99 | 1 | 10 | 2 | 3.821433 | true | main.c | 1 | 142 | {"end_line": 142, "start_line": 1, "target_file": "main.c"} | baseline_code/EX1/histo.zip | 15,570 | a7341b8ad3417fc975416e73f1b78139d1a9b1d8965ca2612157f3f74ad03d62 | 9 | [
"EX1_optimized_codes/main.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/histo | input_data/EX1/histo.zip | 145,074,823 | 7eed5e4252ef272e488a5e6c4ed5f829f1211c7bd933511ea203a443432108cf | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 6,298,638 | parboil | binary | uint8 | .bmp | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as uint8 (little-endian); shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["-i", "input_data/extra-large/input/img.bin", "-o", "/dev/null", "1000"], "large": ["-i", "input_data/large/input/img.bin", "-o", "/dev/null", "1000"], "medium": ["-i", "input_data/medium/input/img.bin", "-o", "/dev/null", "1000"], "mini": ["-i", "input_data/mini/input/img.bin", "-o", "/dev/null", "100... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
hotspot | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Stencil / Scientific Computing | SC | d3 | 106 | 1 | 11 | 6 | 5.986094 | true | hotspot.cpp | 47 | 157 | {"end_line": 157, "start_line": 47, "target_file": "hotspot.cpp"} | baseline_code/EX1/hotspot.zip | 4,505 | 6d78620318a34fc4bc2f1ba534d44479280cfa52b757732ad95cefceae1e9765 | 4 | [
"EX1_optimized_codes/hotspot.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/hotspot | input_data/EX1/hotspot.zip | 456,016,769 | 44f7eb2d8fdf7599c9105a234e928aa1f309bc4191e56aab8de70d4ae9c7fcbc | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 1,413,849,203 | rodinia | text | float64 | .txt | 1 | 1 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1, rtol=1). | {"extra-large": ["8192", "8192", "500", "1", "input_data/extra-large/input/temp_8192", "input_data/extra-large/input/power_8192", "/dev/null"], "large": ["4096", "4096", "500", "1", "input_data/large/input/temp_4096", "input_data/large/input/power_4096", "/dev/null"], "medium": ["1024", "1024", "500", "1", "input_data/... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
hotspot3D | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Stencil / Scientific Computing | SC | d2 | 40 | 1 | 6 | 5 | 3.32277 | true | 3D.c | 127 | 173 | {"end_line": 173, "start_line": 127, "target_file": "3D.c"} | baseline_code/EX1/hotspot3D.zip | 3,640 | 21a78bb9094e5ebf0d126e633936c92f496c2370d9c320456f7431ed220531ac | 3 | [
"EX1_optimized_codes/3D.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/hotspot3D | input_data/EX1/hotspot3D.zip | 116,423,538 | 2dcf156d53f0776ba9c9da8c44802f60651ddc837126c0c40a04ccfe08a765be | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 121,424,000 | rodinia | text | float64 | .txt | 1 | 1 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1, rtol=1). | {"extra-large": ["1024", "4", "500", "input_data/extra-large/input/power_1024x4", "input_data/extra-large/input/temp_1024x4", "/dev/null"], "large": ["512", "8", "500", "input_data/large/input/power_512x8", "input_data/large/input/temp_512x8", "/dev/null"], "medium": ["512", "4", "500", "input_data/medium/input/power_5... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
jacobi-1d | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Stencil / Scientific Computing | SC | d1 | 17 | 1 | 0 | 2 | 0.005507 | true | jacobi-1d.c | 65 | 83 | {"end_line": 83, "start_line": 65, "target_file": "jacobi-1d.c"} | baseline_code/EX1/jacobi-1d.zip | 4,548 | 6376d72e05642366675f2126de3ef05f3900a58d1598059efe2f7f16c321ec30 | 4 | [
"EX1_optimized_codes/jacobi-1d.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/jacobi-1d | input_data/EX1/jacobi-1d.zip | 5,811 | 4322010772fd2c6c48d7098943dce19b28830d656aee86c93af89df2636b3c74 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 33,438 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
jacobi-2d | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Stencil / Scientific Computing | SC | d1 | 19 | 1 | 0 | 4 | 0.928857 | true | jacobi-2d.c | 66 | 86 | {"end_line": 86, "start_line": 66, "target_file": "jacobi-2d.c"} | baseline_code/EX1/jacobi-2d.zip | 4,531 | 8cc0d1dd38b8e4a0a45ae99c7ec66f11d74e04f403d00cac99ba368ba891d1b9 | 4 | [
"EX1_optimized_codes/jacobi-2d.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/jacobi-2d | input_data/EX1/jacobi-2d.zip | 28,650,818 | 7ba6e43c513cc91ac29907601510582fa9057c10166347ade0be753ff1eeaf22 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 67,899,207 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
kmeans | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Clustering / Machine Learning | CML | d2 | 82 | 3 | 3 | 5 | 3.99297 | true | kmeans-clustering.c | 1 | 119 | {"end_line": 119, "start_line": 1, "target_file": "kmeans-clustering.c"} | baseline_code/EX1/kmeans.zip | 30,072 | b191b32f2c23d94e022f919ad8c6e58f909a5793de8ddb2dc5325e1a6469e108 | 13 | [
"EX1_optimized_codes/kmeans-clustering.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/kmeans | input_data/EX1/kmeans.zip | 336,696,259 | 97d886cc3cb5a441bb1ec89541464e88ab06f663db7c4d1bbf9f1564a189ba79 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 6,019 | rodinia | text | float64 | .txt | 0.00001 | 0.00001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-05, rtol=1e-05). | {"extra-large": ["-i", "input_data/extra-large/input/5000000_34.txt", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/1000000_34.txt", "-o", "/dev/null"], "medium": ["-i", "input_data/medium/input/100000_34.txt", "-o", "/dev/null"], "mini": ["-i", "input_data/mini/input/10000_34.txt", "-o", "/dev/null"], "s... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
kmp-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Sorting / Search | SS | d2 | 43 | 4 | 3 | 3 | 2.616854 | true | kmp.c | 1 | 52 | {"end_line": 52, "start_line": 1, "target_file": "kmp.c"} | baseline_code/EX1/kmp-mach.zip | 23,383 | b18dcb68c39d2dbf4ba9f1791d1a2b4e8c92a6ebe450a976e333a900dce1b76d | 10 | [
"EX1_optimized_codes/kmp.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/kmp-mach | input_data/EX1/kmp-mach.zip | 48,436 | f646ccc4c3df24ee19594d8f025e2bd08b3cdd5bb5093cbcf21e305ac6b54292 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 29 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
lavaMD | EX1 | Serial CPU optimization | C | rodinia | Rodinia | N-body Methods | NBM | d1 | 91 | 1 | 1 | 2 | 1.192288 | true | kernel_cpu.c | 1 | 140 | {"end_line": 140, "start_line": 1, "target_file": "kernel_cpu.c"} | baseline_code/EX1/lavaMD.zip | 10,624 | 3addfde137e0a03f1dd6f243e604c3a65f5d2e8edac215774da8a4da496cf497 | 14 | [
"EX1_optimized_codes/kernel_cpu.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/lavaMD | input_data/EX1/lavaMD.zip | 32,015,291 | 92b2ddf68eb352b24238078cfa2c7d5b4cb69ec3b02ef62694701ff68b82a2c3 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 63,471,309 | rodinia | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["-boxes1d", "20", "-o", "/dev/null"], "large": ["-boxes1d", "15", "-o", "/dev/null"], "medium": ["-boxes1d", "10", "-o", "/dev/null"], "mini": ["-boxes1d", "5", "-o", "/dev/null"], "small": ["-boxes1d", "7", "-o", "/dev/null"]} | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
lbm | EX1 | Serial CPU optimization | C | parboil | Parboil | Computational Fluid Dynamics | CFD | d4 | 535 | 14 | 26 | 5 | 9.348315 | true | lbm.c | 1 | 626 | {"end_line": 626, "start_line": 1, "target_file": "lbm.c"} | baseline_code/EX1/lbm.zip | 21,429 | afaadc92a8ecdd465f5d0bca99dac5c46edf4eafceb148380d133e76c83f17e9 | 11 | [
"EX1_optimized_codes/lbm.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/lbm | input_data/EX1/lbm.zip | 93,402,410 | e92cd9eed4df87ea437a2ffde7b7733b5ebda197324cfeb2b3ff8a3d146ad96d | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 129,600,000 | parboil | binary | float32 | .dat | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as float32 (little-endian); shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"extra-large": ["-i", "input_data/extra-large/input/300_300_300_ldc.of", "-o", "/dev/null", "100"], "large": ["-i", "input_data/large/input/240_240_250_ldc.of", "-o", "/dev/null", "100"], "medium": ["-i", "input_data/medium/input/180_180_200_ldc.of", "-o", "/dev/null", "100"], "mini": ["-i", "input_data/mini/input/60_... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
leukocyte | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Image and Video Processing | IVP | d1 | 123 | 1 | 3 | 2 | 2.146369 | true | track-ellipse.c | 4 | 205 | {"end_line": 205, "start_line": 4, "target_file": "track-ellipse.c"} | baseline_code/EX1/leukocyte.zip | 26,789 | 2c624ee4beec1b43341cbe78c3e98a5fa3683366b051f7dc43ec2870a329120c | 14 | [
"EX1_optimized_codes/track-ellipse.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/leukocyte | input_data/EX1/leukocyte.zip | 2,078,179,852 | 3bd2b270bd952a8365f6e768465f875d7e63b7efb05a7013fb76e93f0616f880 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 200 | rodinia | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["150", "4", "input_data/extra-large/testfile.avi", "-o", "/dev/null"], "large": ["100", "4", "input_data/large/testfile.avi", "-o", "/dev/null"], "medium": ["50", "4", "input_data/medium/testfile.avi", "-o", "/dev/null"], "mini": ["5", "4", "input_data/mini/testfile.avi", "-o", "/dev/null"], "small": [... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
lu | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 21 | 1 | 0 | 2 | 0.054454 | true | lu.c | 85 | 106 | {"end_line": 106, "start_line": 85, "target_file": "lu.c"} | baseline_code/EX1/lu.zip | 4,424 | 4c656abce247b8acb1501ecb534fe07c82501ea195912c7916b690b1dae69c0f | 4 | [
"EX1_optimized_codes/lu.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/lu | input_data/EX1/lu.zip | 2,974,274 | 99c1a6ecb4173684bd96fa98f36c1b852be7595110d290e044cece978f6f2ac6 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 101,889,160 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
lud | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Dense Linear Algebra | DLA | d1 | 21 | 1 | 0 | 2 | 0.054454 | true | lud-omp.c | 1 | 23 | {"end_line": 23, "start_line": 1, "target_file": "lud-omp.c"} | baseline_code/EX1/lud.zip | 6,917 | 5a7b5748183bff12a341e69d28c9c9d1c2058c3413a90572270a91380ac51143 | 10 | [
"EX1_optimized_codes/lud-omp.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/lud | input_data/EX1/lud.zip | 104,863,751 | 3bb6e2f0a1280e1204285ef87c49cda56704626f08fe6c70e68213609e01db80 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"large",
"extra-large"
] | 4 | 247,771,800 | rodinia | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["-s", "4096", "-v", "-o", "/dev/null"], "large": ["-s", "3072", "-v", "-o", "/dev/null"], "medium": ["-s", "2048", "-v", "-o", "/dev/null"], "mini": ["-s", "516", "-v", "-o", "/dev/null"], "small": ["-s", "1024", "-v", "-o", "/dev/null"]} | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
ludcmp | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 40 | 1 | 0 | 2 | 0.286955 | true | ludcmp.c | 95 | 139 | {"end_line": 139, "start_line": 95, "target_file": "ludcmp.c"} | baseline_code/EX1/ludcmp.zip | 4,745 | eae85ab74a0d9a936e27e476a7ac24df75e757bfa6fee0fa42593ae9a473f6ab | 4 | [
"EX1_optimized_codes/ludcmp.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/ludcmp | input_data/EX1/ludcmp.zip | 3,895 | 3b54f60d009171116a15f9e47efa9e8d3812be13ff6076456f7bf512b1058ac3 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 39,897 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
matmul | EX1 | Serial CPU optimization | C | codee | Codee | Dense Linear Algebra | DLA | d1 | 20 | 1 | 0 | 2 | 0.042217 | true | matmul.c | 8 | 30 | {"end_line": 30, "start_line": 8, "target_file": "matmul.c"} | baseline_code/EX1/matmul.zip | 4,220 | 5ad44fce20353d0662697e303a1f8d78da2a259c5615235588453d5cc0840fab | 6 | [
"EX1_optimized_codes/matmul.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/matmul | input_data/EX1/matmul.zip | 48,098,823 | 802d33b771b0b47f8fc82f3b6a4d857269fabc6647f86a6d73b8dd1243aaa2d5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 164,912,944 | codee | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["3000", "/dev/null"], "large": ["2000", "/dev/null"], "medium": ["1500", "/dev/null"], "mini": ["500", "/dev/null"], "small": ["1000", "/dev/null"]} | Codee | https://www.codee.com/ | Codee open catalog of benchmarks. |
md-grid-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | N-body Methods | NBM | d2 | 55 | 3 | 3 | 3 | 2.763697 | true | md.c | 1 | 70 | {"end_line": 70, "start_line": 1, "target_file": "md.c"} | baseline_code/EX1/md-grid-mach.zip | 7,396 | a09bf234b277c40b07b456ab9b396f397f915c35aa69aa73fa1881dcc7a4d6fe | 7 | [
"EX1_optimized_codes/md.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/md-grid-mach | input_data/EX1/md-grid-mach.zip | 97,074 | 778f8a0dffd01300ab4bc62ff8e6ee1d922a22f58a5fcbb9254fdecf44403b87 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 184,322 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
md-knn-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | N-body Methods | NBM | d1 | 50 | 3 | 0 | 2 | 1.409325 | true | md.c | 1 | 64 | {"end_line": 64, "start_line": 1, "target_file": "md.c"} | baseline_code/EX1/md-knn-mach.zip | 8,889 | 8ee058c07227d6d9eba1d0f1bb519de1cd947c69d9b154285fdc36312d7398b7 | 8 | [
"EX1_optimized_codes/md.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/md-knn-mach | input_data/EX1/md-knn-mach.zip | 117,656 | 2679fc798dfa36a2675db61cb2f9ff556c788cb37a2f7c04fd76e398437c976e | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 74,913 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
mri-gridding | EX1 | Serial CPU optimization | C | parboil | Parboil | Medical / Scientific Computing | MSC | d3 | 138 | 4 | 8 | 4 | 5.635049 | true | CPU_kernels.c | 1 | 197 | {"end_line": 197, "start_line": 1, "target_file": "CPU_kernels.c"} | baseline_code/EX1/mri-gridding.zip | 17,513 | f5775eb2061d941521c9580d395e332774bdcdd237abf3855e5af5f3a6d3662c | 9 | [
"EX1_optimized_codes/CPU_kernels.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/mri-gridding | input_data/EX1/mri-gridding.zip | 876,971,056 | 51a3248e14d4ec17b610224202a15c21db2bd21c6462ff7033e14efbe6b6f418 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 20 | parboil | binary | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as int32 (little-endian); shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["-i", "input_data/extra-large/input/extra-large.uks", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/large.uks", "-o", "/dev/null"], "medium": ["-i", "input_data/medium/input/medium.uks", "-o", "/dev/null"], "mini": ["-i", "input_data/mini/input/mini.uks", "-o", "/dev/null"], "small": ["-i... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
mri-q | EX1 | Serial CPU optimization | C++ | parboil | Parboil | Medical / Scientific Computing | MSC | d1 | 56 | 3 | 1 | 2 | 1.763996 | true | computeQ.cc | 1 | 65 | {"end_line": 65, "start_line": 1, "target_file": "computeQ.cc"} | baseline_code/EX1/mri-q.zip | 16,675 | 7934053e8d14d0302cec91d47cd019a3958dc4d17d94240774ac4bdf87f2811d | 10 | [
"EX1_optimized_codes/computeQ.cc"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/mri-q | input_data/EX1/mri-q.zip | 14,916,685 | 650f0a90d074b838e414bbff3f924230a9c9dc78ebe22febe29ae4c5c821e9a4 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 7,667,732 | parboil | binary | int32 | .out | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as int32 (little-endian); shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["-i", "input_data/extra-large/input/81_81_81_dataset.bin", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/64_64_64_dataset.bin", "-o", "/dev/null"], "medium": ["-i", "input_data/medium/input/51_51_51_dataset.bin", "-o", "/dev/null"], "mini": ["-i", "input_data/mini/input/20_20_20_dataset.b... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
mvt | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dense Linear Algebra | DLA | d1 | 18 | 1 | 0 | 4 | 0.91662 | true | mvt.c | 79 | 98 | {"end_line": 98, "start_line": 79, "target_file": "mvt.c"} | baseline_code/EX1/mvt.zip | 4,384 | 66e2a54d48ee0a048b99e34e0b6cbd3c87c5d7772e5ead2b5b7aa0bfe8b158e7 | 4 | [
"EX1_optimized_codes/mvt.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/mvt | input_data/EX1/mvt.zip | 38,473 | 873879f2422a3533f0113484de19ac34d6b3d3f193618044fdf47f10ded76ca9 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 95,300 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
myocyte | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Medical / Scientific Computing | MSC | d1 | 62 | 1 | 2 | 2 | 1.118667 | true | master.c | 1 | 85 | {"end_line": 85, "start_line": 1, "target_file": "master.c"} | baseline_code/EX1/myocyte.zip | 27,166 | d52d7948d771d4798d5b5f008bd12c4d399d25934e7ead60671692041e06c6bb | 14 | [
"EX1_optimized_codes/master.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/myocyte | input_data/EX1/myocyte.zip | 6,106 | aa0be88a9d0b4ae5473f8e78d3b12dfb932a1ae877169c976d1f65d214fdbfa2 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 6,050 | rodinia | text | float64 | .txt | 0.000001 | 0.000001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-06, rtol=1e-06). | {"extra-large": ["100", "8000", "0", "4", "-o", "/dev/null"], "large": ["100", "4000", "0", "4", "-o", "/dev/null"], "medium": ["100", "1200", "0", "4", "-o", "/dev/null"], "mini": ["100", "50", "0", "4", "-o", "/dev/null"], "small": ["100", "200", "0", "4", "-o", "/dev/null"]} | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
nussinov | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Dynamic Programming | DP | d2 | 25 | 1 | 4 | 5 | 2.576716 | true | nussinov.c | 81 | 111 | {"end_line": 111, "start_line": 81, "target_file": "nussinov.c"} | baseline_code/EX1/nussinov.zip | 8,499 | 1d33e1f7e881fb4800f7870d7399a329ffb2861ceb541d9e112c16c1c1c84b4e | 5 | [
"EX1_optimized_codes/nussinov.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/nussinov | input_data/EX1/nussinov.zip | 19,467,238 | eb2418b4fba7150002c2b217719810094de730f766035a4f700aa9119f1b3677 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 78,898,681 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
nw | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Dynamic Programming | DP | d3 | 139 | 1 | 17 | 5 | 7.346724 | true | needle.cpp | 133 | 307 | {"end_line": 307, "start_line": 133, "target_file": "needle.cpp"} | baseline_code/EX1/nw.zip | 3,897 | b90ebf568625c5b9ddf1c7d97584c2727f8aae467a50be8fe79d2f771b8073aa | 3 | [
"EX1_optimized_codes/needle.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/nw | input_data/EX1/nw.zip | 3,384 | bed6cc9fad6dc45d9a881e2d7bc205133792021b2449dbaec7d3c1739f4bdb95 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 10 | rodinia | text | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["24576", "10", "1", "-o", "input_data/extra-large/output/output_extra-large.txt"], "large": ["16384", "10", "1", "-o", "input_data/large/output/output_large.txt"], "medium": ["12288", "10", "1", "-o", "input_data/medium/output/output_medium.txt"], "mini": ["4096", "10", "1", "-o", "input_data/mini/outp... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
nw-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Dynamic Programming | DP | d2 | 83 | 3 | 6 | 2 | 3.500642 | true | nw.c | 1 | 104 | {"end_line": 104, "start_line": 1, "target_file": "nw.c"} | baseline_code/EX1/nw-mach.zip | 6,358 | 6f18d158316f8cd1dbe6d07d3aff8b6f4f6bd8734ba226b79bed6b7d7b531c2a | 8 | [
"EX1_optimized_codes/nw.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/nw-mach | input_data/EX1/nw-mach.zip | 12,441 | 5d2c14a41abe3d13df1ca6edfa2c5c9ece5a241614825ff399dc9284f7eb1aea | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 2,615 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
particlefilter | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Monte Carlo Methods | MCM | d2 | 184 | 1 | 7 | 2 | 4.01782 | true | particle-filter.c | 361 | 581 | {"end_line": 581, "start_line": 361, "target_file": "particle-filter.c"} | baseline_code/EX1/particlefilter.zip | 7,286 | 364e7038f30a4915c02a5d2357239e5e5231b5254e9c67d635d9e62b0c126458 | 3 | [
"EX1_optimized_codes/particle-filter.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/particlefilter | input_data/EX1/particlefilter.zip | 3,879 | 511fdabb58ba4d05ff78f77758e739107ee2fa5af227191996a5c94ebeaf1c98 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 250 | rodinia | text | float64 | .txt | 0.00001 | 0.00001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-05, rtol=1e-05). | {"extra-large": ["-x", "1024", "-y", "1024", "-z", "30", "-np", "80000", "-o", "/dev/null"], "large": ["-x", "1024", "-y", "1024", "-z", "20", "-np", "50000", "-o", "/dev/null"], "medium": ["-x", "768", "-y", "768", "-z", "20", "-np", "40000", "-o", "/dev/null"], "mini": ["-x", "512", "-y", "512", "-z", "10", "-np", "2... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
pathfinder | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Dynamic Programming | DP | d2 | 56 | 1 | 6 | 3 | 2.619684 | true | pathfinder.cpp | 104 | 172 | {"end_line": 172, "start_line": 104, "target_file": "pathfinder.cpp"} | baseline_code/EX1/pathfinder.zip | 3,375 | 4c7424c003a501821a93f656e2f5464e35e718d2e80ebdd177e76a005f5fcfb9 | 5 | [
"EX1_optimized_codes/pathfinder.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/pathfinder | input_data/EX1/pathfinder.zip | 4,545,590 | 6b4f11938122401a9c9d809b145540fd74f578aa9cd05c61227e7f198286449a | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 852,000,020 | rodinia | text | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["150000000", "10", "--no-rand", "-o", "/dev/null"], "large": ["100000000", "10", "--no-rand", "-o", "/dev/null"], "medium": ["60000000", "10", "-no-rand", "-o", "/dev/null"], "mini": ["15000000", "10", "-no-rand", "-o", "/dev/null"], "small": ["30000000", "10", "-no-rand", "-o", "/dev/null"]} | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
patricia-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Graph Algorithms | GA | d2 | 72 | 1 | 7 | 4 | 3.546163 | true | patricia.c | 63 | 175 | {"end_line": 175, "start_line": 63, "target_file": "patricia.c"} | baseline_code/EX1/patricia-mibench.zip | 13,453 | 045ec1d54ec0a0641f829adc4efec499377ddbad50f10f8a02503e0cc88e4bc4 | 10 | [
"EX1_optimized_codes/patricia.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/patricia-mibench | input_data/EX1/patricia-mibench.zip | 1,059,077,176 | cf80ae9eb9e11942b06e92548fd095a867c0bb8bb8b9abd3de383db8c2ac597d | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 1,377,227,170 | mibench | text | float64 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"extra-large": ["./input_data/extra-large/input/input.udp"], "large": ["./input_data/large/input/input.udp"], "medium": ["./input_data/medium/input/input.udp"], "mini": ["./input_data/mini/input/input.udp"], "small": ["./input_data/small/input/input.udp"]} | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
pi | EX1 | Serial CPU optimization | C | codee | Codee | Monte Carlo Methods | MCM | d2 | 79 | 2 | 7 | 3 | 3.682383 | true | pi.c | 1 | 103 | {"end_line": 103, "start_line": 1, "target_file": "pi.c"} | baseline_code/EX1/pi.zip | 2,111 | 06a8f1f1840d5ef2691fca9211b488415151c1ce08750c726a7c2f5033412828 | 2 | [
"EX1_optimized_codes/pi.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/pi | input_data/EX1/pi.zip | 3,567 | 9f4678646a45e9bf04f4649cc32b2ee4fe99afadbd3fc1c6ce3e5f46297edcc9 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 285 | codee | text | float64 | .txt | 0.00001 | 0.00001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-05, rtol=1e-05). | {"extra-large": ["2000000000", "/dev/null"], "large": ["1200000000", "/dev/null"], "medium": ["800000000", "/dev/null"], "mini": ["200000000", "/dev/null"], "small": ["400000000", "/dev/null"]} | Codee | https://www.codee.com/ | Codee open catalog of benchmarks. |
qsort-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Sorting / Search | SS | d3 | 94 | 2 | 11 | 4 | 5.440375 | true | qsort_large.c | 1 | 114 | {"end_line": 114, "start_line": 1, "target_file": "qsort_large.c"} | baseline_code/EX1/qsort-mibench.zip | 12,749,673 | ffafa5eb702ea33ec820f2cfe34a5cb9f457d8b64be0a157a810b7c7b0bff693 | 10 | [
"EX1_optimized_codes/qsort_large.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/qsort-mibench | input_data/EX1/qsort-mibench.zip | 1,437,781,932 | 5737f0de4dd61ff741bbfe309a245a863e9f84ada2d58f8c8b1c8177bbefc4b5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 1,395,686,010 | mibench | text | int32 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as int32 tokens; shapes must match; values compared with exact equality (numpy.array_equal). | {"extra-large": ["./input_data/extra-large/input/input_extra-large.dat", "/dev/null"], "large": ["./input_data/large/input/input_large.dat", "/dev/null"], "medium": ["./input_data/medium/input/input_medium.dat", "/dev/null"], "mini": ["./input_data/mini/input/input_mini.dat", "/dev/null"], "small": ["./input_data/small... | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
sad | EX1 | Serial CPU optimization | C | parboil | Parboil | Image and Video Processing | IVP | d3 | 158 | 3 | 4 | 8 | 5.85591 | true | sad_cpu.c | 1 | 202 | {"end_line": 202, "start_line": 1, "target_file": "sad_cpu.c"} | baseline_code/EX1/sad.zip | 22,481 | 1050cc5030107967e464bb8fda604d436745e20edc0dd78a7dd5126c49730b3d | 13 | [
"EX1_optimized_codes/sad_cpu.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/sad | input_data/EX1/sad.zip | 30,064,033 | 0b08ad8265e7d636d6fe20016be129b4f26afba8889590acc61df9a35c49d7b5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 246,998,308 | parboil | binary | uint16 | .bin | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as uint16 (little-endian); shapes must match; values compared with exact equality (numpy.array_equal). | {"default": ["-i", "input_data/mini/input/reference.bin,input_data/mini/input/frame.bin", "-o", "/dev/null"], "extra-large": ["-i", "input_data/extra-large/input/reference.bin,input_data/extra-large/input/frame.bin", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/reference.bin,input_data/large/input/frame.... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
seidel-2d | EX1 | Serial CPU optimization | C | polybench | PolyBench/C | Stencil / Scientific Computing | SC | d1 | 15 | 1 | 0 | 3 | 0.449438 | true | seidel-2d.c | 62 | 78 | {"end_line": 78, "start_line": 62, "target_file": "seidel-2d.c"} | baseline_code/EX1/seidel-2d.zip | 4,371 | e65db89a796ecb2218491f642248dab40de70acbb05ebcd7d9bb1bb101110994 | 4 | [
"EX1_optimized_codes/seidel-2d.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/seidel-2d | input_data/EX1/seidel-2d.zip | 60,588,806 | 9829c1bd903195d4de0e5d9bb6517db4c94ab53cd26acf39fc1294f756537c44 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 146,111,157 | polybench | text | float64 | .txt | 0.002 | 0 | benchmark catalog (scripts/benchmark_catalog.py) | Text dump of the output arrays; numeric tokens are compared element-wise with numpy.isclose(candidate, reference, atol=0.002, rtol=0). | null | PolyBench/C | https://www.cs.colostate.edu/~pouchet/software/polybench/ | L.-N. Pouchet. PolyBench/C: The Polyhedral Benchmark Suite. |
sgemm | EX1 | Serial CPU optimization | C++ | parboil | Parboil | Dense Linear Algebra | DLA | d1 | 22 | 1 | 2 | 3 | 1.078629 | true | sgemm_kernel.cc | 1 | 24 | {"end_line": 24, "start_line": 1, "target_file": "sgemm_kernel.cc"} | baseline_code/EX1/sgemm.zip | 15,378 | 10eaa4ed2e51e20d837645b211d5b0f6e45b1e43c4876ff8c0588a3cd139611e | 8 | [
"EX1_optimized_codes/sgemm_kernel.cc"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/sgemm | input_data/EX1/sgemm.zip | 289,965,399 | 281bd54dae7471c40394759c8c365bb3065ba09416459273c65678404e7c860b | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 59,372,969 | parboil | text | float32 | .txt | 0.00001 | 0.00001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float32 tokens; shapes must match; values compared with numpy.allclose(atol=1e-05, rtol=1e-05). | {"extra-large": ["-i", "input_data/extra_large/input/matrix1.txt,input_data/extra_large/input/matrix2.txt,input_data/extra_large/input/matrix2t.txt", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/matrix1.txt,input_data/large/input/matrix2.txt,input_data/large/input/matrix2t.txt", "-o", "/dev/null"], "medi... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
sort-merge-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Sorting / Search | SS | d1 | 52 | 4 | 2 | 2 | 1.996298 | true | sort.c | 1 | 64 | {"end_line": 64, "start_line": 1, "target_file": "sort.c"} | baseline_code/EX1/sort-merge-mach.zip | 5,522 | bac31d17eaaada4942e4809c8789e13093aa18bd3413111543270acf6d1ec4a3 | 7 | [
"EX1_optimized_codes/sort.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/sort-merge-mach | input_data/EX1/sort-merge-mach.zip | 116,846 | 115c9325d77b6bf8e2cc216eeff468a19d9dcef455200e3222d04d9e87bae2cd | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 107,315 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
sort-radix-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Sorting / Search | SS | d1 | 29 | 1 | 2 | 2 | 0.714849 | true | sort.c | 77 | 110 | {"end_line": 110, "start_line": 77, "target_file": "sort.c"} | baseline_code/EX1/sort-radix-mach.zip | 6,648 | 5bf54c3566bfb6fbdd8b9c7684017b2d532c996399caeb09e3a5ea289120a566 | 7 | [
"EX1_optimized_codes/sort.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/sort-radix-mach | input_data/EX1/sort-radix-mach.zip | 117,180 | b2652f20a6793cc24183f565ec88046a8a46de7a597665739f824eff76d1109f | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 107,315 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
spmv | EX1 | Serial CPU optimization | C | parboil | Parboil | Sparse Linear Algebra | SLA | d1 | 102 | 1 | 5 | 2 | 2.451894 | true | main.c | 1 | 134 | {"end_line": 134, "start_line": 1, "target_file": "main.c"} | baseline_code/EX1/spmv.zip | 26,134 | 085abeeea492e0545b767c430b737d828f58c0b90f693025f986a1de5ac21f70 | 15 | [
"EX1_optimized_codes/main.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/spmv | input_data/EX1/spmv.zip | 72,523,416 | 5a7c39a0820407c94cddee357b4eb53f91d613d11eb87558d2eec071bdbf4ef7 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large"
] | 4 | 1,278,740 | parboil | binary | float32 | .out | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as float32 (little-endian); shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"extra-large": ["-i", "input_data/extra-large/input/matrix.mtx,input_data/extra-large/input/vector.bin", "-o", "/dev/null"], "large": ["-i", "input_data/large/input/Dubcova3.mtx.bin,input_data/large/input/vector.bin", "-o", "/dev/null"], "medium": ["-i", "input_data/medium/input/bcsstk18.mtx,input_data/medium/input/ve... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
spmv-crs-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Sparse Linear Algebra | SLA | d1 | 24 | 3 | 0 | 2 | 1.091165 | true | spmv.c | 1 | 30 | {"end_line": 30, "start_line": 1, "target_file": "spmv.c"} | baseline_code/EX1/spmv-crs-mach.zip | 7,866 | fa5d5aa9df3584a81ec5c542b3da10398fbc5a38dd2c1aa877ce363ad0d610be | 8 | [
"EX1_optimized_codes/spmv.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/spmv-crs-mach | input_data/EX1/spmv-crs-mach.zip | 125,506 | da32cfb5aa556b5ffe5ecf265294556fd5af2c7d77a9062779aa6189b85c9536 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 50,878 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
spmv-ellpack-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Sparse Linear Algebra | SLA | d1 | 23 | 3 | 0 | 2 | 1.078928 | true | spmv.c | 1 | 29 | {"end_line": 29, "start_line": 1, "target_file": "spmv.c"} | baseline_code/EX1/spmv-ellpack-mach.zip | 7,451 | 8ddf6735a82e6390d1332446759769db2ba2921978bec67907673ad80ac6d008 | 7 | [
"EX1_optimized_codes/spmv.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/spmv-ellpack-mach | input_data/EX1/spmv-ellpack-mach.zip | 141,121 | 2f023c5b916863b0c1159e6b4ab70fb7c91ce0e4f92dac1d4584af92bc411460 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 50,878 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
srad_v1 | EX1 | Serial CPU optimization | C | rodinia | Rodinia | Stencil / Scientific Computing | SC | d2 | 239 | 1 | 6 | 2 | 4.1875 | true | srad.c | 1 | 343 | {"end_line": 343, "start_line": 1, "target_file": "srad.c"} | baseline_code/EX1/srad_v1.zip | 8,269 | d66484af033a2f5bcfe8edd7fd80bb63241ab1bd884a251ab90a958b3f675886 | 9 | [
"EX1_optimized_codes/srad.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/srad_v1 | input_data/EX1/srad_v1.zip | 11,150,998 | 0ce5b007e5467cecc0d96486d2d93053cdb66766b6445a5a883fd5754897f592 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 181,179,605 | rodinia | text | int32 | .pgm | 0 | 0 | output specification (scripts/checkers/output_specs.py) | PGM image; pixel arrays must be exactly equal (numpy.array_equal). | {"default": ["100", "0.5", "502", "458", "1", "/dev/null"], "extra-large": ["250", "0.5", "6024", "5496", "1", "/dev/null"], "large": ["200", "0.5", "4016", "3664", "1", "/dev/null"], "medium": ["150", "0.5", "2008", "1832", "1", "/dev/null"], "mini": ["100", "0.5", "502", "458", "1", "/dev/null"], "small": ["100", "0.... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
srad_v2 | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Stencil / Scientific Computing | SC | d3 | 199 | 3 | 10 | 3 | 6.494562 | true | srad.cpp | 1 | 248 | {"end_line": 248, "start_line": 1, "target_file": "srad.cpp"} | baseline_code/EX1/srad_v2.zip | 3,980 | d693c62d61da333955c0841280458a82ac5fc33e1be01ec5216ed27a0da5b2a3 | 4 | [
"EX1_optimized_codes/srad.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/srad_v2 | input_data/EX1/srad_v2.zip | 109,523,865 | 3332743c1b41a9d990e400f4bfc7600dc463992b390081a7589700421cb9275f | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 245,825,536 | rodinia | text | float64 | .txt | 0.1 | 0.1 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=0.1, rtol=0.1). | {"extra-large": ["4096", "4096", "0", "1023", "0", "1023", "8", "0.5", "250", "/dev/null"], "large": ["2048", "2048", "0", "511", "0", "511", "8", "0.5", "200", "/dev/null"], "medium": ["1024", "1024", "0", "255", "0", "255", "8", "0.5", "150", "/dev/null"], "mini": ["256", "256", "0", "63", "0", "63", "8", "0.5", "100... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
stencil | EX1 | Serial CPU optimization | C | parboil | Parboil | Stencil / Scientific Computing | SC | d1 | 22 | 1 | 0 | 2 | 0.066691 | true | kernels.c | 1 | 25 | {"end_line": 25, "start_line": 1, "target_file": "kernels.c"} | baseline_code/EX1/stencil.zip | 17,995 | c76f1a31836978f6cfc5d92e05c897673a9024059d6c8202005cb3382f27021a | 11 | [
"EX1_optimized_codes/kernels.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/stencil | input_data/EX1/stencil.zip | 470,107,963 | ea3bca5d19c344b364ac2e34f8c67a000d564063ed992acf72e9e32256d5448d | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 309,329,940 | parboil | binary | float32 | .out | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Binary file read as float32 (little-endian); shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"extra-large": ["-i", "input_data/extra-large/input/768x768x96.bin", "-o", "/dev/null", "768", "768", "96", "100"], "large": ["-i", "input_data/large/input/512x512x64x100.bin", "-o", "/dev/null", "512", "512", "64", "100"], "medium": ["-i", "input_data/medium/input/256x256x48.bin", "-o", "/dev/null", "256", "256", "48... | Parboil | http://impact.crhc.illinois.edu/parboil/parboil.aspx | J. A. Stratton et al. Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing. IMPACT-12-01, 2012. |
stencil-2d-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Stencil / Scientific Computing | SC | d1 | 26 | 3 | 0 | 2 | 1.115639 | true | stencil.c | 1 | 32 | {"end_line": 32, "start_line": 1, "target_file": "stencil.c"} | baseline_code/EX1/stencil-2d-mach.zip | 5,352 | 4c905d6e2b321c77fc4f274ac7d56c454ca0365cb8abaf89f2912aec32299bb0 | 6 | [
"EX1_optimized_codes/stencil.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/stencil-2d-mach | input_data/EX1/stencil-2d-mach.zip | 242,605 | 602f1c0511e467df496a958a07d5f6df710fa3e9e4e58be82410ce3adb47b563 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 315,666 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
stencil-3d-mach | EX1 | Serial CPU optimization | C | machsuite | MachSuite | Stencil / Scientific Computing | SC | d1 | 48 | 3 | 0 | 2 | 1.384851 | true | stencil.c | 1 | 58 | {"end_line": 58, "start_line": 1, "target_file": "stencil.c"} | baseline_code/EX1/stencil-3d-mach.zip | 5,782 | 146fe2758a809ab2bf57b905dbcbbc7a1a761180571baa56ed8652adee8b923c | 6 | [
"EX1_optimized_codes/stencil.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/stencil-3d-mach | input_data/EX1/stencil-3d-mach.zip | 356,113 | 9485dd699986834fb6330a43e6a7da3e8d23c9cdd39240bd660358cf426e1dba | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 397,043 | machsuite | text | null | .data | null | null | program self-check (upstream check.data) | Self-checking program: it compares its result against input_data/<size>/output/check.data (upstream MachSuite reference) and prints Success/Fail; the harness scans captured stdout. | {"default": ["./input_data/mini/input/input.data", "./input_data/mini/output/check.data"], "extra-large": ["./input_data/extra-large/input/input.data", "./input_data/extra-large/output/check.data"], "large": ["./input_data/large/input/input.data", "./input_data/large/output/check.data"], "medium": ["./input_data/medium... | MachSuite | https://github.com/breagen/MachSuite | B. Reagen et al. MachSuite: Benchmarks for Accelerator Design and Customized Architectures. IISWC 2014. |
streamcluster | EX1 | Serial CPU optimization | C++ | rodinia | Rodinia | Clustering / Machine Learning | CML | d3 | 170 | 1 | 22 | 3 | 6.827192 | true | streamcluster.cpp | 354 | 597 | {"end_line": 597, "start_line": 354, "target_file": "streamcluster.cpp"} | baseline_code/EX1/streamcluster.zip | 11,716 | 474f6caca9c5c53183ad6b2e96d842fc2f60bdabb7a32f14284949e0eae00e4c | 4 | [
"EX1_optimized_codes/streamcluster.cpp"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/streamcluster | input_data/EX1/streamcluster.zip | 22,310 | b6ceb715dd07db8f1baafd738a1e55d408a779c08c169ee8e8df7dee9aee9fa8 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 43,528 | rodinia | text | float64 | .txt | 0.00001 | 0.00001 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=1e-05, rtol=1e-05). | {"extra-large": ["10", "60", "128", "98304", "98304", "500", "none", "/dev/null", "1"], "large": ["10", "40", "128", "65536", "65536", "500", "none", "/dev/null", "1"], "medium": ["10", "20", "64", "16384", "16384", "200", "none", "/dev/null", "1"], "mini": ["10", "20", "32", "1024", "1024", "100", "none", "/dev/null",... | Rodinia | https://www.cs.virginia.edu/rodinia/ | S. Che et al. Rodinia: A Benchmark Suite for Heterogeneous Computing. IISWC 2009. |
stringsearch-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Sorting / Search | SS | d1 | 30 | 1 | 2 | 3 | 1.176525 | true | pbmsrch_parameterized.c | 26 | 57 | {"end_line": 57, "start_line": 26, "target_file": "pbmsrch_parameterized.c"} | baseline_code/EX1/stringsearch-mibench.zip | 17,799 | fc638067b1d47cc5a54a27df619e52ae506b52c4793a2672258a31788f16db61 | 13 | [
"EX1_optimized_codes/pbmsrch_parameterized.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/stringsearch-mibench | input_data/EX1/stringsearch-mibench.zip | 3,000,121,935 | 4e1ede80d9a787f1672a3aaf4bcc148e6099b6ba121262617fcb5f4ee7c8941c | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 21,839,101,386 | mibench | text | float64 | .txt | 0 | 0 | output specification (scripts/checkers/output_specs.py) | Text output parsed as float64 tokens; shapes must match; values compared with numpy.allclose(atol=0, rtol=0). | {"extra-large": ["extra-large"], "large": ["large"], "medium": ["medium"], "mini": ["mini"], "small": ["small"]} | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
susan-c-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Image and Video Processing | IVP | d4 | 253 | 1 | 26 | 74 | 9 | true | susan.c | 1,184 | 1,469 | {"end_line": 1469, "start_line": 1184, "target_file": "susan.c"} | baseline_code/EX1/susan-c-mibench.zip | 13,444 | 6358d00ee109942b46e9b966a8e2ea02baf841eaf1af667dfdb2c72ad420c630 | 5 | [
"EX1_optimized_codes/susan.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/susan-c-mibench | input_data/EX1/susan-c-mibench.zip | 521,892,504 | 37c1e6e61825a39006eadcde8e04e0814148f5ef77238d6f4cb6272b5abd1097 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"large",
"extra-large"
] | 4 | 254,811,254 | mibench | binary | uint8 | .pgm | 0 | 0 | output specification (scripts/checkers/output_specs.py) | PGM image; pixel arrays must be exactly equal (numpy.array_equal). | {"extra-large": ["./input_data/extra-large/input/input_extra-large.pgm", "/dev/null", "-c"], "large": ["./input_data/large/input/input_large.pgm", "/dev/null", "-c"], "medium": ["./input_data/medium/input/input_medium.pgm", "/dev/null", "-c"], "mini": ["./input_data/mini/input/input_mini.pgm", "/dev/null", "-c"], "smal... | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
susan-e-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Image and Video Processing | IVP | d4 | 206 | 1 | 16 | 4 | 7.717158 | true | susan.c | 783 | 1,023 | {"end_line": 1023, "start_line": 783, "target_file": "susan.c"} | baseline_code/EX1/susan-e-mibench.zip | 11,320 | c8d2735c05a706a3c6607bc103dead0ced017fb8bedc54ad7113ef1b912f721b | 4 | [
"EX1_optimized_codes/susan.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/susan-e-mibench | input_data/EX1/susan-e-mibench.zip | 370,037,200 | 976007dde6a2983ac75cb7935600a2d1e6ef6ee5a23e1c3cedbc4c8ce7b10d6f | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 267,394,183 | mibench | binary | uint8 | .pgm | 0 | 0 | output specification (scripts/checkers/output_specs.py) | PGM image; pixel arrays must be exactly equal (numpy.array_equal). | {"extra-large": ["./input_data/extra-large/input/input_extra-large.pgm", "/dev/null", "-e"], "large": ["./input_data/large/input/input_large.pgm", "/dev/null", "-e"], "medium": ["./input_data/medium/input/input_medium.pgm", "/dev/null", "-e"], "mini": ["./input_data/mini/input/input_mini.pgm", "/dev/null", "-e"], "smal... | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
susan-s-mibench | EX1 | Serial CPU optimization | C | mibench | MiBench | Image and Video Processing | IVP | d2 | 108 | 1 | 6 | 4 | 3.705442 | true | susan.c | 391 | 529 | {"end_line": 529, "start_line": 391, "target_file": "susan.c"} | baseline_code/EX1/susan-s-mibench.zip | 23,960 | a3b9552b28b68b6386c993a067bef1dfaba51f18a305c74b6d9f7932255136ed | 9 | [
"EX1_optimized_codes/susan.c"
] | https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench | 7dc9824ba10249babe9c898b7591415b79445e22 | EX1/susan-s-mibench | input_data/EX1/susan-s-mibench.zip | 531,824,160 | 7ef731939f66c49e1e034cd85da4d63b8bf481448fbe3d834b632a71d35193ba | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | [
"mini",
"small",
"medium",
"large",
"extra-large"
] | 5 | 267,394,183 | mibench | binary | uint8 | .pgm | 0 | 0 | output specification (scripts/checkers/output_specs.py) | PGM image; pixel arrays must be exactly equal (numpy.array_equal). | {"default": ["./input_data/mini/input/input_mini.pgm", "/dev/null", "-s"], "extra-large": ["./input_data/extra-large/input/input_extra-large.pgm", "/dev/null", "-s"], "large": ["./input_data/large/input/input_large.pgm", "/dev/null", "-s"], "medium": ["./input_data/medium/input/input_medium.pgm", "/dev/null", "-s"], "m... | MiBench | https://vhosts.eecs.umich.edu/mibench/ | M. R. Guthaus et al. MiBench: A Free, Commercially Representative Embedded Benchmark Suite. WWC 2001. |
HPC-Bench
Benchmark data for HPC-Bench: A Comprehensive Benchmark for High Performance Computing Codes (NeurIPS 2026 Evaluations and Datasets Track).
HPC-Bench evaluates large language models as optimizers of computational kernels embedded in complete HPC programs. Each task is a profiler-identified kernel region of a real benchmark program; a candidate replaces only that region, and the whole program is compiled, run on standardized inputs, checked for correctness against a reference output, and timed. The suite covers 92 CPU workloads evaluated under serial (EX1) and OpenMP (EX2) optimization and 22 Rodinia-derived CUDA workloads evaluated under GPU optimization (EX3), spanning 16 computational motifs.
This dataset holds everything needed to run the benchmark that is not source code of the evaluation pipeline: the standardized inputs at five scales, the reference outputs used for correctness checking, the exact instrumented baseline programs, and a structured index that ties them together. The evaluation pipeline, the LLM-generated candidates studied in the paper, and the per-trial results live in the GitHub repository.
- Code and pipeline: https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench
- Paper: https://arxiv.org/abs/2611.02752
Contents
| Component | Path | Size | Notes |
|---|---|---|---|
| Input data, 5 scales per benchmark | input_data/{EX1,EX2,EX3}/<benchmark>.zip |
207 archives, 37.7 GB | One archive per setting and benchmark; also contains the reference outputs |
| Reference (golden) outputs | inside the input archives, input_data/<size>/output/ |
1,023 files, 78.9 GB uncompressed | Indexed in index/golden_outputs.parquet; not duplicated |
| Instrumented baseline source | baseline_code/{EX1,EX2,EX3}/<benchmark>.zip |
208 archives, 35 MB | Exact baseline programs used in the experiments; no LLM-generated code |
| Shared build infrastructure | baseline_code/build_infrastructure.zip |
190 files | common.mk, per-suite Makefile infrastructure, PolyBench utilities |
| Benchmark index | index/benchmarks.parquet |
208 rows | One row per (setting, benchmark) |
| Input-size index | index/inputs.parquet |
1,040 rows | One row per (setting, benchmark, input size) |
| Reference-output index | index/golden_outputs.parquet |
1,023 rows | One row per reference-output file |
| Archive manifests | manifest.csv, baseline_code/manifest.csv |
Path, size, and SHA-256 of every archive (index/input_archives.parquet is a Parquet copy of manifest.csv for the Dataset Viewer) |
|
| Croissant metadata | croissant.json |
Croissant 1.1 with NeurIPS RAI fields |
Settings: EX1 serial CPU optimization (gcc -O3), EX2 OpenMP CPU parallelization (gcc -O3 -fopenmp, thread count set by OMP_NUM_THREADS), EX3 CUDA GPU optimization (nvcc). EX1 and EX2 share the same 92 benchmarks and the same inputs. EX3 contains 24 benchmark directories: the 22 CUDA benchmarks reported in the paper, plus backprop and nw, which have CUDA sources and inputs but are not part of the paper's CUDA benchmark table (in_paper_benchmark_set = false). nn generates its input at run time and has no input archive.
Layout
bcui2/HPC-Bench/
βββ README.md
βββ croissant.json
βββ manifest.csv # input archives: ex, benchmark, path, bytes, sha256
βββ input_data/
β βββ EX1|EX2|EX3/<benchmark>.zip # -> input_data/<size>/{input,output}/...
βββ baseline_code/
β βββ manifest.csv
β βββ build_infrastructure.zip # common.mk, common_*/, driver_parboil/, utilities/, EX*/Makefile
β βββ EX1|EX2|EX3/<benchmark>.zip # -> <benchmark>/... (Makefile, sources, EX*_optimized_codes/<baseline kernel>)
βββ index/
βββ benchmarks.parquet
βββ inputs.parquet
βββ golden_outputs.parquet
Each input archive unpacks to a single input_data/ directory with one subdirectory per scale (mini, small, medium, large, extra-large). Inputs are under input_data/<size>/input/ (PolyBench kernels generate their inputs in the program and have no input files) and reference outputs under input_data/<size>/output/.
Quick start
Inputs and reference outputs
Clone the GitHub repository and let its fetch script download, verify, and extract the archives into EX*/<benchmark>/input_data/:
git clone https://github.com/Deep-Learning-Profiling-Tools/HPC-Bench.git
cd HPC-Bench
python tools/fetch_input_data.py # everything (~38 GB zipped)
python tools/fetch_input_data.py --ex EX1 --benchmarks 2mm,bfs
python tools/fetch_input_data.py --list
The script reads manifest.csv from this dataset, checks each archive's SHA-256, and extracts it in place. Reference outputs arrive with the inputs.
Index
import pandas as pd
b = pd.read_parquet("hf://datasets/bcui2/HPC-Bench/index/benchmarks.parquet")
b[b.setting == "EX1"][["benchmark", "suite", "motif_abbrev", "structural_complexity", "checker", "atol", "rtol"]]
or with datasets: load_dataset("bcui2/HPC-Bench", "benchmarks") (also "inputs", "golden_outputs", "input_archives").
Baseline source code
from huggingface_hub import hf_hub_download
import zipfile
for f in ["baseline_code/build_infrastructure.zip", "baseline_code/EX1/2mm.zip"]:
p = hf_hub_download("bcui2/HPC-Bench", f, repo_type="dataset")
zipfile.ZipFile(p).extractall("EX1" if "/EX1/" in f else ".")
baseline_code/EX1/2mm.zip unpacks to 2mm/..., so extracting it inside EX1/ recreates EX1/2mm/ as it is in the GitHub repository (minus the LLM candidates). build_infrastructure.zip unpacks at the repository root. The benchmark Makefiles include the shared infrastructure through relative paths, so keep this layout. Building and running are driven by scripts/arg.py in the GitHub repository:
python scripts/arg.py benchmark build --root . --benchmarks 2mm --ex-versions EX1
Benchmark index (index/benchmarks.parquet)
One row per (setting, benchmark): 92 EX1, 92 EX2, 24 EX3.
| Column group | Columns |
|---|---|
| Identity | benchmark, setting (EX1/EX2/EX3), setting_name, language (C, C++, CUDA), suite, suite_name |
| Classification | motif, motif_abbrev (16 motifs), structural_complexity (d1βd4), complexity_loc, complexity_functions, complexity_branches, complexity_loop_depth, complexity_score, in_paper_benchmark_set |
| Annotated region | annotated_target_file, annotated_start_line, annotated_end_line (the kernel region the model may rewrite; annotated_regions_json keeps multi-region cases verbatim) |
| Baseline code | baseline_code_path, baseline_code_bytes, baseline_code_sha256, baseline_code_files, baseline_kernel_files, source_repo, source_commit, source_path |
| Inputs | input_archive_path, input_archive_bytes, input_archive_sha256, input_sizes, n_input_sizes |
| Reference outputs | golden_output_sizes, n_golden_output_sizes, golden_output_bytes |
| Correctness check | checker, output_type, output_dtype, output_suffix, atol, rtol, tolerance_source, comparison_rule |
| Run configuration | runtime_args_json (command-line arguments per input size, from configs/benchmark_args.json) |
| Provenance | upstream_suite, upstream_url, upstream_reference |
structural_complexity is the paper's static complexity level computed from the annotated region (branch count, lines of code, loop depth, function count). The raw metrics are given for the CPU benchmarks; for EX3 the level comes from the paper's CUDA benchmark table and the raw metrics are not released.
index/inputs.parquet has one row per (setting, benchmark, size) with size_present, input_files, input_bytes, golden_output_available, golden_output_files (paths inside the archive), golden_output_bytes, and runtime_args. index/golden_outputs.parquet has one row per reference-output file with its path_in_archive, bytes, and the comparison parameters of its benchmark.
Reference outputs and correctness checking
A candidate is correct when the program's output, after the candidate kernel replaces the annotated region, matches the reference output for the same input under the suite's comparison rule. The reference output for each input is the output of the instrumented baseline program on that input, stored at input_data/<size>/output/ inside the input archive; MachSuite ships its own check.data reference that the program compares against itself. Both are consumed in place by the checkers in scripts/checkers/ of the GitHub repository.
| Suite | Checker | Reference | Comparison rule |
|---|---|---|---|
| PolyBench | polybench |
text dump of the output arrays | numeric tokens compared element-wise, numpy.isclose(atol=atol, rtol=0); atol is 0.002 (0.0005 for one benchmark) |
| Rodinia, Parboil, Codee, MiBench | suite checker | text or binary file (output_type, output_dtype, output_suffix) |
shapes must match; floating-point data numpy.allclose(atol=atol, rtol=rtol), integer data and PGM images exactly equal |
| MachSuite | machsuite |
check.data from the upstream suite |
the program compares its own result against check.data and prints Success/Fail; the captured stdout is scanned |
PARSEC (canneal) |
parsec |
text summary | the candidate's Improvement: value must be at least the reference value (absolute tolerance 1e-9, relative 1e-6) |
atol and rtol are the tolerances the checker applies when correctness is checked with scripts/arg.py benchmark check, and tolerance_source says where they are defined: the PolyBench checker uses the per-benchmark threshold in scripts/benchmark_catalog.py; the Rodinia, Parboil, Codee, and MiBench checkers use the per-benchmark output specification in scripts/checkers/output_specs.py; the PARSEC checker uses constants in scripts/checkers/parsec.py. Integer outputs and images are compared exactly (atol = rtol = 0), and MachSuite rows carry no tolerance because the program checks itself. comparison_rule states the rule in words for every row.
Four CUDA-only benchmarks (dwt2d, gaussian, huffman, hybridsort) have reference outputs in their archives but no registered output specification, so the released checker framework records their checks as skipped and their candidates are not counted as correct in the released EX3 results; they are marked checker = unregistered, as is nn, which has no reference output.
Coverage: of the 1,016 (setting, benchmark, size) combinations present in the archives, 1,004 carry reference outputs. The exceptions are lud medium (EX1, EX2), spmv extra-large (EX1, EX2) and mini (EX2), susan-c-mibench medium (EX1, EX2), and all sizes of EX3 srad_v2; haccmk (EX1, EX2) and EX3 nn generate their data inside the program and have no per-size directories, and EX3 backprop, bilateral, and gaussian ship fewer than five scales. index/inputs.parquet records this per row.
To regenerate a reference output, build the baseline with scripts/arg.py benchmark build as above, run the baseline executable for that size with the arguments in runtime_args, and capture its output (stdout for text outputs, or the file the program writes) to input_data/<size>/output/. The baseline source, compiler flags, and inputs released here are the ones used for the published references; a different compiler or architecture can change floating-point results within the stated tolerances.
Baseline source code
baseline_code/<EX>/<benchmark>.zip contains the benchmark directory from the GitHub repository at commit 7dc9824ba10249babe9c898b7591415b79445e22, restricted to the baseline: Makefile, program sources and headers, data generators, and the baseline kernel file(s) in <EX>_optimized_codes/ (baseline_kernel_files in the index). All LLM-generated candidate files (*_<model>_v<N>.*, 6,464 files) and all input data are excluded. The kernels carry the timing instrumentation used for the paper's kernel-time measurements; the annotated optimization region is given by annotated_target_file and the line range in the index. Every archive is listed with its SHA-256 in baseline_code/manifest.csv.
Benchmark composition
| Suite | EX1/EX2 workloads | Upstream |
|---|---|---|
| PolyBench/C | 30 | https://www.cs.colostate.edu/~pouchet/software/polybench/ |
| Rodinia | 19 (+ CUDA set) | https://www.cs.virginia.edu/rodinia/ |
| MachSuite | 18 | https://github.com/breagen/MachSuite |
| Parboil | 10 | http://impact.crhc.illinois.edu/parboil/parboil.aspx |
| MiBench | 9 | https://vhosts.eecs.umich.edu/mibench/ |
| Codee | 5 | https://www.codee.com/ |
| PARSEC | 1 | https://parsec.cs.princeton.edu/ |
Motifs (EX1/EX2 counts): Dense Linear Algebra 22, Stencil / Scientific Computing 13, Image and Video Processing 7, Graph Algorithms 6, N-body Methods 6, Dynamic Programming 5, Sorting / Search 5, Clustering / Machine Learning 4, Medical / Scientific Computing 4, Statistical Computations 4, Utility Benchmarks 4, Monte Carlo Methods 3, Sparse Linear Algebra 3, Cryptography / Encoding 2, Computational Fluid Dynamics 2, Spectral Methods 2. Structural complexity: 61 d1, 17 d2, 9 d3, 5 d4.
Intended use and limitations
HPC-Bench is intended for evaluating code-generating models on correctness-preserving performance optimization of HPC kernels under serial, OpenMP, and CUDA settings, for studying the failure modes of such optimizations, and for testing compilers, autotuners, and program-optimization tools against fixed inputs and references. It measures kernel-level optimization of profiled regions, not end-to-end application performance or general software-engineering ability.
The benchmarks are kernels embedded in complete but moderately sized programs from academic suites; libraries such as BLAS or MPI are not exercised, and all programs are C, C++, or CUDA. Coverage is skewed toward dense linear algebra and stencil kernels and toward structurally simple regions (61 of 92 CPU workloads are d1); sparse linear algebra is covered by three SpMV kernels. The reference outputs and tolerances were established with GCC 14.2 and CUDA 12.4 at -O3 on an AMD EPYC 7543 / NVIDIA A100 system; other compilers or architectures can change floating-point results within the stated tolerances. The data contain no personal or sensitive information; inputs are synthetic or standard scientific datasets.
Reproducibility
The published evaluation runs entirely from this dataset plus the GitHub repository: tools/fetch_input_data.py places the inputs and reference outputs, baseline_code/ provides the exact baseline programs at a fixed commit, the index gives the annotated region, run arguments, and comparison rule for every benchmark, and scripts/arg.py in the repository builds, runs, checks, and times candidates. The LLM candidates and per-trial results behind the paper's figures are in the repository (EX*/<benchmark>/EX*_optimized_codes/, final_data/), with a figure-by-figure guide in docs/REPRODUCING_PAPER.md.
Licensing and attribution
The benchmark programs and input data are derived from PolyBench/C, Rodinia, Parboil, MachSuite, MiBench, PARSEC, and Codee and remain subject to the licenses and attribution requirements of those suites; the upstream references are given per row in the index (upstream_suite, upstream_url, upstream_reference) and in the GitHub repository's docs/BENCHMARKS.md. The Codee kernels (atmux, coulomb, haccmk, matmul, pi) are redistributed with Codee's explicit permission, which extends to continued redistribution after the conclusion of the research collaboration in which they were obtained. The index files, manifests, metadata, and the HPC-Bench instrumentation and tooling are released under the MIT License by the HPC-Bench authors. Users who redistribute benchmark programs or inputs should check the license of the originating suite.
Citation
@article{cui2026hpc,
title={HPC-Bench: A Comprehensive Benchmark for High Performance Computing Codes},
author={Cui, Bowen and Yin, Junyu and Ramesh, Tejas and Lim, Leo and Hernandez, Oscar and Zhou, Keren},
journal={arXiv preprint arXiv:2611.02752},
year={2026}
}
Croissant
croissant.json describes this dataset in Croissant 1.1, including the Responsible AI fields required by the NeurIPS 2026 Evaluations and Datasets Track, and was validated with mlcroissant.
- Downloads last month
- 199