Hugging Face's logo

Datasets:
uralstech
/
AIDE-Chip-15K-gem5-Sims

Tasks:
Tabular Regression
Modalities:
Tabular
Text
Formats:
csv
Languages:
English
Size:
10K - 100K
Tags:
computer-architecture
gem5
cache
design-space-exploration
surrogate-modeling
microarchitecture
DOI:
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
Community
1
Dataset Viewer
Auto-converted to Parquet Duplicate
l1d_size
int64
16
128
l1i_size
int64
16
128
l2_size
int64
128
1.02k
l1d_assoc
int64
1
8
l1i_assoc
int64
1
8
l2_assoc
int64
1
16
workload
stringclasses
6 values
ipc
float64
0.22
4.09
l2_miss_rate
float64
0
1
sim_duration_s
float64
10.4
1.73k
error
int64
0
0
error_msg
null
64
32
256
1
2
16
matrix_mul
0.808749
0.073967
735.24
0
null
128
32
256
8
1
16
matrix_mul
0.839999
0.06191
741.52
0
null
16
32
256
1
8
16
matrix_mul
0.798548
0.078226
742.48
0
null
16
32
256
4
2
4
matrix_mul
0.951992
0.025958
750.97
0
null
64
16
256
8
4
16
matrix_mul
0.798833
0.078449
753.25
0
null
16
16
256
8
8
16
matrix_mul
0.799269
0.078548
754.79
0
null
128
16
256
2
2
4
matrix_mul
0.958637
0.023921
757.73
0
null
64
32
256
2
2
16
matrix_mul
0.806867
0.074966
759.17
0
null
32
128
256
1
4
4
matrix_mul
0.951586
0.025954
764.76
0
null
32
64
256
8
8
4
matrix_mul
0.951336
0.026027
764.95
0
null
128
64
256
4
1
16
matrix_mul
0.913307
0.03658
765.49
0
null
16
16
256
2
1
16
matrix_mul
0.799071
0.078453
766.26
0
null
16
128
256
2
8
2
matrix_mul
0.989469
0.017639
768.62
0
null
128
128
256
2
1
1
matrix_mul
1.012201
0.013094
776.01
0
null
32
64
256
4
2
16
matrix_mul
0.798951
0.078509
778.91
0
null
128
64
256
8
2
2
matrix_mul
0.996459
0.0158
780.77
0
null
64
32
512
1
1
2
matrix_mul
0.997535
0.01127
781.02
0
null
32
64
256
1
2
2
matrix_mul
0.988515
0.017687
785.44
0
null
64
16
512
8
1
2
matrix_mul
1.001263
0.010823
786.18
0
null
128
32
256
4
1
16
matrix_mul
0.913305
0.03658
790.89
0
null
32
32
512
1
2
2
matrix_mul
0.99583
0.011622
796.64
0
null
64
64
512
1
8
2
matrix_mul
0.997544
0.01127
799.54
0
null
16
64
512
2
2
16
matrix_mul
1.023844
0.001495
806.39
0
null
32
64
1,024
8
1
16
matrix_mul
1.026971
0.000762
808.16
0
null
32
16
512
1
4
16
matrix_mul
1.022005
0.001493
809.95
0
null
32
32
512
2
4
4
matrix_mul
1.023543
0.001487
811.4
0
null
32
128
512
4
8
16
matrix_mul
1.024319
0.001495
814.08
0
null
32
64
512
2
2
16
matrix_mul
1.024286
0.001495
815.83
0
null
16
32
512
8
4
8
matrix_mul
1.023834
0.00149
822.43
0
null
32
128
1,024
2
2
8
matrix_mul
1.027
0.000762
823.1
0
null
16
16
512
8
1
4
matrix_mul
1.023571
0.001487
823.34
0
null
16
64
1,024
2
8
4
matrix_mul
1.026585
0.000762
823.93
0
null
32
128
1,024
2
1
16
matrix_mul
1.026999
0.000762
826.19
0
null
64
64
512
2
8
1
matrix_mul
0.999538
0.007853
826.42
0
null
128
16
512
4
8
8
matrix_mul
1.023981
0.00149
828.16
0
null
128
128
512
4
2
8
matrix_mul
1.023986
0.00149
830.81
0
null
32
128
1,024
1
1
8
matrix_mul
1.021397
0.00076
832.23
0
null
16
32
512
1
4
16
matrix_mul
1.019604
0.001488
832.29
0
null
64
16
512
4
4
8
matrix_mul
1.023912
0.00149
835
0
null
32
32
1,024
8
8
16
matrix_mul
1.026975
0.000762
838.52
0
null
64
128
512
2
1
16
matrix_mul
1.024346
0.001495
838.61
0
null
16
64
1,024
4
1
1
matrix_mul
1.026953
0.000762
839.26
0
null
16
128
512
1
1
16
matrix_mul
1.019603
0.001488
845.16
0
null
64
64
512
8
4
1
matrix_mul
1.000344
0.00765
853.6
0
null
128
32
1,024
1
8
16
matrix_mul
1.025141
0.000761
854.91
0
null
32
16
1,024
4
8
4
matrix_mul
1.027013
0.000762
872.82
0
null
128
128
1,024
8
8
4
matrix_mul
1.027204
0.000762
873.12
0
null
16
16
256
2
2
1
matrix_mul
1.009722
0.013591
751.96
0
null
16
128
256
4
1
1
matrix_mul
1.010226
0.013546
759.78
0
null
64
128
128
8
1
1
matrix_mul
0.222332
0.999984
1,533.89
0
null
16
128
256
8
2
16
matrix_mul
0.799272
0.078548
736
0
null
16
32
128
1
8
8
matrix_mul
0.222168
0.996172
1,559.46
0
null
128
32
128
2
1
4
matrix_mul
0.222352
0.999972
1,568.68
0
null
16
64
128
4
4
16
matrix_mul
0.222338
0.999802
1,569.77
0
null
32
16
128
4
4
4
matrix_mul
0.22234
0.999865
1,577.16
0
null
32
64
128
8
4
8
matrix_mul
0.222339
0.99985
1,583.44
0
null
16
128
1,024
8
8
8
matrix_mul
1.027007
0.000762
819.2
0
null
32
16
1,024
1
1
8
matrix_mul
1.02139
0.00076
807.58
0
null
16
64
128
2
2
4
matrix_mul
0.222338
0.999836
1,588.95
0
null
16
128
256
2
2
4
matrix_mul
0.951183
0.026079
765.07
0
null
64
16
512
1
8
4
matrix_mul
1.022248
0.001485
831.58
0
null
128
32
1,024
8
4
8
matrix_mul
1.027204
0.000762
833.49
0
null
32
16
512
4
4
1
matrix_mul
0.998478
0.008132
814.69
0
null
128
16
256
4
8
8
matrix_mul
0.915793
0.036982
776.42
0
null
16
32
512
8
2
4
matrix_mul
1.023584
0.001487
836.25
0
null
128
128
256
1
4
1
matrix_mul
1.008775
0.013722
768.94
0
null
64
16
512
4
8
8
matrix_mul
1.0239
0.00149
812.42
0
null
128
32
512
2
4
16
matrix_mul
1.024407
0.001495
818.86
0
null
128
128
1,024
1
8
4
matrix_mul
1.025141
0.000761
829.8
0
null
128
16
128
4
1
1
matrix_mul
0.222357
0.999997
1,606.49
0
null
128
64
512
2
4
1
matrix_mul
1.00209
0.007169
825.55
0
null
16
128
256
2
4
16
matrix_mul
0.799081
0.078453
735.13
0
null
128
64
256
1
8
4
matrix_mul
0.96831
0.021847
757.68
0
null
64
32
256
2
1
8
matrix_mul
0.898266
0.042823
785.7
0
null
64
16
128
8
8
16
matrix_mul
0.222343
0.999939
1,617.52
0
null
32
32
1,024
1
4
1
matrix_mul
1.021399
0.00076
803.31
0
null
64
128
256
8
2
8
matrix_mul
0.898558
0.042667
809.67
0
null
16
128
1,024
2
1
4
matrix_mul
1.026584
0.000762
861.85
0
null
32
16
1,024
2
1
2
matrix_mul
1.026967
0.000763
801.1
0
null
128
128
1,024
1
4
1
matrix_mul
1.025141
0.000761
846.28
0
null
32
128
512
2
4
4
matrix_mul
1.023543
0.001487
827.9
0
null
128
16
512
4
2
4
matrix_mul
1.023746
0.001488
835.96
0
null
64
128
512
1
4
8
matrix_mul
1.022544
0.001487
826.81
0
null
32
32
1,024
4
1
8
matrix_mul
1.027009
0.000762
836.69
0
null
64
16
1,024
8
8
1
matrix_mul
1.027073
0.000762
855.82
0
null
32
32
1,024
8
1
4
matrix_mul
1.026969
0.000762
882.95
0
null
32
64
1,024
4
4
8
matrix_mul
1.027015
0.000762
866.74
0
null
128
32
512
8
2
16
matrix_mul
1.024499
0.001496
833.24
0
null
128
16
256
4
1
16
matrix_mul
0.9133
0.03658
775.84
0
null
64
32
256
4
8
2
matrix_mul
0.990027
0.017614
753.7
0
null
16
16
128
4
4
16
matrix_mul
0.222334
0.999801
1,572.25
0
null
32
32
256
2
1
4
matrix_mul
0.951439
0.026034
781.95
0
null
64
128
128
1
4
2
matrix_mul
0.222205
0.999293
1,562.95
0
null
128
32
128
2
8
1
matrix_mul
0.222344
1
1,589.83
0
null
64
128
128
4
2
8
matrix_mul
0.222343
0.999924
1,579.4
0
null
16
64
1,024
2
4
8
matrix_mul
1.026585
0.000762
830.99
0
null
16
128
128
1
1
1
matrix_mul
0.222012
0.997256
1,601.37
0
null
16
16
256
1
4
16
matrix_mul
0.798505
0.078226
746.17
0
null
32
128
128
4
8
2
matrix_mul
0.222335
0.999926
1,512.17
0
null
128
128
256
1
4
8
matrix_mul
0.96485
0.023112
746.92
0
null
End of preview. Expand in Data Studio

AIDE-Chip 15K gem5 Simulation Dataset

AIDE-Chip-15K-gem5-Sims is a structured dataset of approximately 15,000 validated RISC-V gem5 simulations covering cache hierarchy design-space exploration (DSE) for single-core processors.

The dataset was generated using gem5's Syscall Emulation (SE) mode and six representative workloads, spanning compute-bound, memory-bound, and irregular access patterns. Each sample maps cache configuration parameters to IPC and L2 miss rate, enabling training of fast, physically consistent surrogate models.

This dataset accompanies the paper:

Udayshankar Ravikumar . Fast, Explainable Surrogate Models for gem5 Cache Design Space Exploration. Authorea. January 14, 2026. https://doi.org/10.22541/au.176843174.46109183/v1

Supported Tasks

  • Cache performance regression (IPC)
  • Cache miss-rate regression (L2 miss rate)

Workloads

Workload Description
crc32 Streaming, low locality
dijkstra Pointer-chasing, irregular
fft Strided, cache-sensitive
matrix_mul Dense compute, high reuse
qsort Branchy, mixed locality
sha Compute-bound, near-zero miss rate

The C code of the workloads can be found at: https://github.com/Uralstech/AIDE-Chip-Surrogates/tree/main/15k-Sims/benchmarks

Dataset Structure

The dataset is released as sharded CSV files for scalability.

Each row contains:

Column Description
l1d_size, l1i_size, l2_size Cache sizes (KB)
l1d_assoc, l1i_assoc, l2_assoc Associativities
workload Benchmark name
ipc Instructions per cycle
l2_miss_rate L2 miss rate
sim_duration_s gem5 simulation wall time
error Simulation success flag
error_msg Simulation error message

Generation Details

  • Simulator: gem5 (SE mode)
  • Execution platform:
    • 4× AWS c6g + 4× AWS c7g
    • 64 vCPUs each
  • Sampling strategy:
    • Constrained grid over cache sizes & associativities
    • Validity constraints enforced
    • Randomized execution order
  • Recommended dataset split:
    • 70% train / 15% validation / 15% test (per workload)

The script used to generate the configuration set can be found at: https://github.com/Uralstech/AIDE-Chip-Surrogates/blob/main/15k-Sims/config-gen/generate_configs.py

Intended Use

This dataset is intended for:

  • Research on surrogate modeling for architecture simulation
  • Cache design-space exploration
  • Explainable ML for systems
  • Educational and academic use

Not intended for commercial use (see License).

Patent Notice

This dataset accompanies research describing surrogate-based techniques for microarchitectural design-space exploration.

The author has filed a pending patent application that may cover broader system-level methods beyond the specific data provided here.

This notice is informational only and does not alter the dataset’s Creative Commons (CC BY-NC-SA 4.0) license.

Downloads last month
16
Size of downloaded dataset files:
758 kB
Size of the auto-converted Parquet files:
319 kB
Number of rows:
15,000

Models trained or fine-tuned on uralstech/AIDE-Chip-15K-gem5-Sims

Space using uralstech/AIDE-Chip-15K-gem5-Sims 1