Datasets:
Remove predictions and roofline_quadrant configs, keep 3 configs only
Browse files
README.md
CHANGED
|
@@ -28,14 +28,6 @@ configs:
|
|
| 28 |
data_files:
|
| 29 |
- split: train
|
| 30 |
path: kernel_profiles/kernels_labeled.parquet
|
| 31 |
-
- config_name: roofline_quadrant
|
| 32 |
-
data_files:
|
| 33 |
-
- split: train
|
| 34 |
-
path: kernel_profiles/roofline_quadrant.parquet
|
| 35 |
-
- config_name: predictions
|
| 36 |
-
data_files:
|
| 37 |
-
- split: train
|
| 38 |
-
path: predictions/serving_predictions.parquet
|
| 39 |
dataset_info:
|
| 40 |
- config_name: trace_replay
|
| 41 |
features:
|
|
@@ -226,121 +218,11 @@ dataset_info:
|
|
| 226 |
splits:
|
| 227 |
- name: train
|
| 228 |
num_examples: 148077
|
| 229 |
-
- config_name: roofline_quadrant
|
| 230 |
-
features:
|
| 231 |
-
- name: model
|
| 232 |
-
dtype: string
|
| 233 |
-
- name: profile
|
| 234 |
-
dtype: string
|
| 235 |
-
- name: concurrency
|
| 236 |
-
dtype: int64
|
| 237 |
-
- name: engine
|
| 238 |
-
dtype: string
|
| 239 |
-
- name: hardware
|
| 240 |
-
dtype: string
|
| 241 |
-
- name: oi
|
| 242 |
-
dtype: float64
|
| 243 |
-
- name: cf_gb
|
| 244 |
-
dtype: float64
|
| 245 |
-
- name: output_tput
|
| 246 |
-
dtype: float64
|
| 247 |
-
- name: tpot_ms
|
| 248 |
-
dtype: float64
|
| 249 |
-
- name: ttft_ms
|
| 250 |
-
dtype: float64
|
| 251 |
-
splits:
|
| 252 |
-
- name: train
|
| 253 |
-
num_examples: 2163
|
| 254 |
-
- config_name: predictions
|
| 255 |
-
features:
|
| 256 |
-
- name: hardware_config
|
| 257 |
-
dtype: string
|
| 258 |
-
- name: model
|
| 259 |
-
dtype: string
|
| 260 |
-
- name: backend
|
| 261 |
-
dtype: string
|
| 262 |
-
- name: profile
|
| 263 |
-
dtype: string
|
| 264 |
-
- name: data_scope
|
| 265 |
-
dtype: string
|
| 266 |
-
- name: concurrency
|
| 267 |
-
dtype: int64
|
| 268 |
-
- name: isl
|
| 269 |
-
dtype: int64
|
| 270 |
-
- name: osl
|
| 271 |
-
dtype: int64
|
| 272 |
-
- name: calibration_status
|
| 273 |
-
dtype: string
|
| 274 |
-
- name: ttft_validation_scope
|
| 275 |
-
dtype: string
|
| 276 |
-
- name: ttft_kernel_ms
|
| 277 |
-
dtype: float64
|
| 278 |
-
- name: ttft_base_ms
|
| 279 |
-
dtype: float64
|
| 280 |
-
- name: ttft_floor_ms
|
| 281 |
-
dtype: float64
|
| 282 |
-
- name: ttft_first_decode_ms
|
| 283 |
-
dtype: float64
|
| 284 |
-
- name: ttft_queue_ms
|
| 285 |
-
dtype: float64
|
| 286 |
-
- name: itl_meas
|
| 287 |
-
dtype: float64
|
| 288 |
-
- name: total_context_tokens
|
| 289 |
-
dtype: int64
|
| 290 |
-
- name: new_prefill_tokens
|
| 291 |
-
dtype: int64
|
| 292 |
-
- name: cached_context_tokens
|
| 293 |
-
dtype: int64
|
| 294 |
-
- name: cache_hit_rate
|
| 295 |
-
dtype: float64
|
| 296 |
-
- name: cache_aware_applied
|
| 297 |
-
dtype: bool
|
| 298 |
-
- name: cache_feature_source
|
| 299 |
-
dtype: string
|
| 300 |
-
- name: cache_prediction_regime
|
| 301 |
-
dtype: string
|
| 302 |
-
- name: ttft_prediction_supported
|
| 303 |
-
dtype: bool
|
| 304 |
-
- name: multiturn_prediction_mode
|
| 305 |
-
dtype: string
|
| 306 |
-
- name: predicted_turn_count
|
| 307 |
-
dtype: float64
|
| 308 |
-
- name: total_successful_turn_requests
|
| 309 |
-
dtype: float64
|
| 310 |
-
- name: mean_predicted_turn_ttft_ms
|
| 311 |
-
dtype: float64
|
| 312 |
-
- name: mean_predicted_turn_tpot_ms
|
| 313 |
-
dtype: float64
|
| 314 |
-
- name: multiturn_turn_predictions
|
| 315 |
-
dtype: string
|
| 316 |
-
- name: ttft_pred
|
| 317 |
-
dtype: float64
|
| 318 |
-
- name: ttft_meas
|
| 319 |
-
dtype: float64
|
| 320 |
-
- name: ttft_err
|
| 321 |
-
dtype: float64
|
| 322 |
-
- name: tpot_pred
|
| 323 |
-
dtype: float64
|
| 324 |
-
- name: tpot_meas
|
| 325 |
-
dtype: float64
|
| 326 |
-
- name: tpot_err
|
| 327 |
-
dtype: float64
|
| 328 |
-
- name: e2el_pred
|
| 329 |
-
dtype: float64
|
| 330 |
-
- name: e2el_meas
|
| 331 |
-
dtype: float64
|
| 332 |
-
- name: e2el_err
|
| 333 |
-
dtype: float64
|
| 334 |
-
- name: measurement_semantics_warning
|
| 335 |
-
dtype: string
|
| 336 |
-
splits:
|
| 337 |
-
- name: train
|
| 338 |
-
num_examples: 4715
|
| 339 |
---
|
| 340 |
|
| 341 |
# AgentPerfBench
|
| 342 |
|
| 343 |
-
LLM inference benchmark: 3,392 serving runs
|
| 344 |
|
| 345 |
## Dataset configurations
|
| 346 |
|
|
@@ -360,14 +242,6 @@ ISL/OSL sampled from lognormal fits to real workload statistics. 42 combinations
|
|
| 360 |
|
| 361 |
Per-kernel Nsight Compute (ncu) profiles across 4 GPUs (A100, H100, RTX 3090, RTX 2080 Ti) and 13 model/sweep sources.
|
| 362 |
|
| 363 |
-
### roofline_quadrant (2,163 rows)
|
| 364 |
-
|
| 365 |
-
Operational intensity and achieved throughput per kernel, for roofline analysis. H100 reference hardware (989 peak TFLOPS, 3.35 TB/s HBM).
|
| 366 |
-
|
| 367 |
-
### predictions (4,715 rows)
|
| 368 |
-
|
| 369 |
-
Predicted vs. measured TTFT, TPOT, and E2EL for each serving configuration, with cache-aware prediction metadata. 14 hardware configs.
|
| 370 |
-
|
| 371 |
### Concurrency filtering
|
| 372 |
|
| 373 |
Concurrency is controlled by a fixed-size connection pool. Trace replay uses levels {1, 5, 10, 20, 40, 80}; distributional uses {1, 5, 10, 40, 80, 200, 320}.
|
|
@@ -449,7 +323,7 @@ Each row in `summary.parquet` (both configs):
|
|
| 449 |
from datasets import load_dataset
|
| 450 |
|
| 451 |
ds = load_dataset("agent-perf-bench/AgentPerfBench", "trace_replay")
|
| 452 |
-
# or "distributional", "kernels_labeled"
|
| 453 |
```
|
| 454 |
|
| 455 |
## Benchmark methodology
|
|
|
|
| 28 |
data_files:
|
| 29 |
- split: train
|
| 30 |
path: kernel_profiles/kernels_labeled.parquet
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
dataset_info:
|
| 32 |
- config_name: trace_replay
|
| 33 |
features:
|
|
|
|
| 218 |
splits:
|
| 219 |
- name: train
|
| 220 |
num_examples: 148077
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 221 |
---
|
| 222 |
|
| 223 |
# AgentPerfBench
|
| 224 |
|
| 225 |
+
LLM inference benchmark: 3,392 serving runs and 148,077 per-kernel NCU profiles across 9 models, 14 GPU configurations, and 2 serving engines (vLLM 0.19.0, SGLang 0.5.9). All models served in BF16 except gpt-oss, which uses mxfp4 for projection weights.
|
| 226 |
|
| 227 |
## Dataset configurations
|
| 228 |
|
|
|
|
| 242 |
|
| 243 |
Per-kernel Nsight Compute (ncu) profiles across 4 GPUs (A100, H100, RTX 3090, RTX 2080 Ti) and 13 model/sweep sources.
|
| 244 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 245 |
### Concurrency filtering
|
| 246 |
|
| 247 |
Concurrency is controlled by a fixed-size connection pool. Trace replay uses levels {1, 5, 10, 20, 40, 80}; distributional uses {1, 5, 10, 40, 80, 200, 320}.
|
|
|
|
| 323 |
from datasets import load_dataset
|
| 324 |
|
| 325 |
ds = load_dataset("agent-perf-bench/AgentPerfBench", "trace_replay")
|
| 326 |
+
# or "distributional", "kernels_labeled"
|
| 327 |
```
|
| 328 |
|
| 329 |
## Benchmark methodology
|