search

mlcommons / inference_results_v4.1

This repository contains the results and code for the MLPerf™ Inference v4.1 benchmark.
https://mlcommons.org/benchmarks/inference-datacenter/
Apache License 2.0
2 stars 7 forks source link

NVIDIA SDXL "Detected system did not match any known systems." #5

Open zixianwang2022 opened 1 day ago

zixianwang2022 commented 1 day ago

Hi I am running into this issue. Are there any ways that I can run the SDXL benchmark on an Intel cpu? 

(mlperf) ziw081@mlperf-inference-ziw081-x86-64-10725:/work$ make run_harness RUN_ARGS="--benchmarks=stable-diffusion-xl --scenarios=Offline --test_mode=PerformanceOnly"
[2024-10-19 19:36:27,381 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC0. Skipping.
[2024-10-19 19:36:27,381 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC1. Skipping.
[2024-10-19 19:36:27,381 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC2. Skipping.
[2024-10-19 19:36:27,381 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC3. Skipping.
[2024-10-19 19:36:27,382 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC4. Skipping.
[2024-10-19 19:36:27,382 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC5. Skipping.
[2024-10-19 19:36:27,382 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC6. Skipping.
[2024-10-19 19:36:27,382 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC7. Skipping.
[2024-10-19 19:36:27,382 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC8. Skipping.
[2024-10-19 19:36:27,382 systems.py:197 INFO] Found unknown device in GPU connection topology: NIC9. Skipping.
[2024-10-19 19:36:27,625 main.py:227 INFO] Detected system did not match any known systems. Exiting. SystemConfiguration(host_cpu_conf=CPUConfiguration(layout={CPU(name='Intel(R) Xeon(R) Platinum 8480+', architecture=<CPUArchitecture.x86_64: AliasedName(name='x86_64', aliases=(), patterns=())>, core_count=56, threads_per_core=2): 2}), host_mem_conf=MemoryConfiguration(host_memory_capacity=Memory(quantity=2.1134187, byte_suffix=<ByteSuffix.TB: (1000, 4)>, _num_bytes=2113418700000), comparison_tolerance=0.05), accelerator_conf=AcceleratorConfiguration(layout=defaultdict(<class 'int'>, {GPU(name='NVIDIA H100 80GB HBM3', accelerator_type=<AcceleratorType.Discrete: AliasedName(name='Discrete', aliases=(), patterns=())>, vram=Memory(quantity=79.6474609375, byte_suffix=<ByteSuffix.GiB: (1024, 3)>, _num_bytes=85520809984), max_power_limit=700.0, pci_id='0x233010DE', compute_sm=90): 8})), numa_conf=NUMAConfiguration(numa_nodes={}, num_numa_nodes=2), system_id=None)

======================== Result summaries: ========================
arjunsuresh commented 1 day ago

Hi @zixianwang2022. Nvidia implementation won't run on Intel CPUs. You can use the reference implementation for it.