search

vllm-project / vllm

A high-throughput and memory-efficient inference and serving engine for LLMs
https://docs.vllm.ai
Apache License 2.0
29.74k stars 4.49k forks source link

[Bug]: In SamplingParams, setting n to a large value (e.g., 512) never finishes #6646

Open RylanSchaeffer opened 3 months ago

RylanSchaeffer commented 3 months ago

Your current environment

$ python collect_env.py
Collecting environment information...
PyTorch version: 2.3.0
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 20.04.6 LTS (x86_64)
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
Clang version: Could not collect
CMake version: version 3.16.3
Libc version: glibc-2.31

Python version: 3.11.5 (main, Sep 11 2023, 13:54:46) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.4.0-162-generic-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA A100-SXM4-80GB
GPU 1: NVIDIA A100-SXM4-80GB
GPU 2: NVIDIA A100-SXM4-80GB
GPU 3: NVIDIA A100-SXM4-80GB
GPU 4: NVIDIA A100-SXM4-80GB
GPU 5: NVIDIA A100-SXM4-80GB
GPU 6: NVIDIA A100-SXM4-80GB
GPU 7: NVIDIA A100-SXM4-80GB

Nvidia driver version: 535.54.03
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                       x86_64
CPU op-mode(s):                     32-bit, 64-bit
Byte Order:                         Little Endian
Address sizes:                      48 bits physical, 48 bits virtual
CPU(s):                             128
On-line CPU(s) list:                0-127
Thread(s) per core:                 2
Core(s) per socket:                 32
Socket(s):                          2
NUMA node(s):                       2
Vendor ID:                          AuthenticAMD
CPU family:                         25
Model:                              1
Model name:                         AMD EPYC 7543 32-Core Processor
Stepping:                           1
Frequency boost:                    enabled
CPU MHz:                            1479.981
CPU max MHz:                        2800.0000
CPU min MHz:                        1500.0000
BogoMIPS:                           5599.76
Virtualization:                     AMD-V
L1d cache:                          2 MiB
L1i cache:                          2 MiB
L2 cache:                           32 MiB
L3 cache:                           512 MiB
NUMA node0 CPU(s):                  0-31,64-95
NUMA node1 CPU(s):                  32-63,96-127
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit:        Not affected
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Not affected
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Not affected
Vulnerability Retbleed:             Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Retpolines, IBPB conditional, IBRS_FW, STIBP always-on, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected
Flags:                              fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 invpcid_single hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload vgif umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca

Versions of relevant libraries:
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.26.4
[pip3] torch==2.3.0
[pip3] torchaudio==2.3.0
[pip3] torchvision==0.18.0
[pip3] transformers==4.40.2
[pip3] triton==2.3.0
[conda] blas                      1.0                         mkl  
[conda] ffmpeg                    4.3                  hf484d3e_0    pytorch
[conda] libjpeg-turbo             2.0.0                h9bf148f_0    pytorch
[conda] mkl                       2023.1.0         h213fc3f_46344  
[conda] mkl-service               2.4.0           py311h5eee18b_1  
[conda] mkl_fft                   1.3.8           py311h5eee18b_0  
[conda] mkl_random                1.2.4           py311hdb19cb5_0  
[conda] numpy                     1.26.4          py311h08b1b3b_0  
[conda] numpy-base                1.26.4          py311hf175353_0  
[conda] pytorch                   2.3.0           py3.11_cuda12.1_cudnn8.9.2_0    pytorch
[conda] pytorch-cuda              12.1                 ha16c6d3_5    pytorch
[conda] pytorch-mutex             1.0                        cuda    pytorch
[conda] torchaudio                2.3.0               py311_cu121    pytorch
[conda] torchtriton               2.3.0                     py311    pytorch
[conda] torchvision               0.18.0              py311_cu121    pytorch
[conda] transformers              4.40.2                   pypi_0    pypi
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: N/A
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X  NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-31,64-95  0       N/A
GPU1    NV12     X  NV12    NV12    NV12    NV12    NV12    NV12    0-31,64-95  0       N/A
GPU2    NV12    NV12     X  NV12    NV12    NV12    NV12    NV12    0-31,64-95  0       N/A
GPU3    NV12    NV12    NV12     X  NV12    NV12    NV12    NV12    0-31,64-95  0       N/A
GPU4    NV12    NV12    NV12    NV12     X  NV12    NV12    NV12    32-63,96-127    1       N/A
GPU5    NV12    NV12    NV12    NV12    NV12     X  NV12    NV12    32-63,96-127    1       N/A
GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X  NV12    32-63,96-127    1       N/A
GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X  32-63,96-127    1       N/A

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks

🐛 Describe the bug

For a research project, I need to generate a large number of outputs per prompt. If I set n in SamplingParams() to be higher than 256, the process hangs indefinitely.

Code to reproduce:

from vllm import LLM, SamplingParams

model = LLM(
    model="EleutherAI/pythia-2.8b",
)

prompts = ["What is the meaning of life?" for _ in range(100)]

for n in [8, 16, 32, 64, 128, 256, 512, 1024, 2048, 4096]:
    model.generate(
        prompts=prompts,
        sampling_params=SamplingParams(n=n, max_tokens=150),
    )
    print(f"Successfully generated outputs for n={n}")

The above script will successfully generate output for up to n=256, but then nothing will happen. VRAM will continue to be occupied, but GPU utilization will be 0:

+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.54.03              Driver Version: 535.54.03    CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA A100-SXM4-80GB          On  | 00000000:07:00.0 Off |                    0 |
| N/A   33C    P0              71W / 350W |  74547MiB / 81920MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+

The process has been like this for 2 days. No error message has been output.

Rosenberg37 commented 1 month ago

+1, 512 hangs, while 256 fine. increase swap_space to 16GB not help.

Rosenberg37 commented 1 month ago

+1, 512 hangs, while 256 fine. increase swap_space to 16GB not help.

I later manager to get through it by further increase the swap_space to the 32. llm = LLM(model=model_path, swap_space=32) I guess the short of swap_space can be cause. @RylanSchaeffer

RylanSchaeffer commented 1 month ago

My solution was to loop with smaller n. It works well enough for my purposes, but I wish a warning would be thrown. Hanging doesn't seem like an acceptable outcome, to me.

youkaichao commented 1 month ago

vllm is designed for fast inference and serving, and it is not designed for scaling test time compute :) I think you need to have an outer loop calling vllm, and scale the test time compute via scaling the outer loop.

RylanSchaeffer commented 1 month ago

@youkaichao

  1. I don't know why your comment is relevant. I was trying to increase n for best-of-n for serving.

  2. Even if your point is relevant, the process shouldn't hang indefinitely. A warning or error should be thrown.