[Bug]: Non-torch memory tracking fails to account for gpu usage of other processes #18854
Description
Your current environment
==============================
System Info
==============================
OS : Red Hat Enterprise Linux 9.5 (Plow) (x86_64)
GCC version : (GCC) 11.5.0 20240719 (Red Hat 11.5.0-5)
Clang version : Could not collect
CMake version : Could not collect
Libc version : glibc-2.34
==============================
PyTorch Info
==============================
PyTorch version : 2.7.0+cu126
Is debug build : False
CUDA used to build PyTorch : 12.6
ROCM used to build PyTorch : N/A
==============================
Python Environment
==============================
Python version : 3.12.5 (main, Apr 2 2025, 00:00:00) [GCC 11.5.0 20240719 (Red Hat 11.5.0-5)] (64-bit runtime)
Python platform : Linux-5.14.0-284.88.1.el9_2.x86_64-x86_64-with-glibc2.34
==============================
CUDA / GPU Info
==============================
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
Nvidia driver version : 550.127.08
cuDNN version : Could not collect
HIP runtime version : N/A
MIOpen runtime version : N/A
Is XNNPACK available : True
==============================
CPU Info
==============================
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 46 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 80
On-line CPU(s) list: 0-79
Vendor ID: GenuineIntel
Model name: Intel Xeon Processor (Cascadelake)
CPU family: 6
Model: 85
Thread(s) per core: 2
Core(s) per socket: 20
Socket(s): 2
Stepping: 6
BogoMIPS: 4799.99
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology cpuid tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves arat umip pku ospke avx512_vnni md_clear arch_capabilities
Virtualization: VT-x
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 2.5 MiB (80 instances)
L1i cache: 2.5 MiB (80 instances)
L2 cache: 160 MiB (40 instances)
L3 cache: 32 MiB (2 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0-39
NUMA node1 CPU(s): 40-79
Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Mitigation; Enhanced IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
==============================
Versions of relevant libraries
==============================
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.26.4
[pip3] nvidia-cublas-cu12==12.6.4.1
[pip3] nvidia-cuda-cupti-cu12==12.6.80
[pip3] nvidia-cuda-nvrtc-cu12==12.6.77
[pip3] nvidia-cuda-runtime-cu12==12.6.77
[pip3] nvidia-cudnn-cu12==9.5.1.17
[pip3] nvidia-cufft-cu12==11.3.0.4
[pip3] nvidia-cufile-cu12==1.11.1.6
[pip3] nvidia-curand-cu12==10.3.7.77
[pip3] nvidia-cusolver-cu12==11.7.1.2
[pip3] nvidia-cusparse-cu12==12.5.4.2
[pip3] nvidia-cusparselt-cu12==0.6.3
[pip3] nvidia-nccl-cu12==2.26.2
[pip3] nvidia-nvjitlink-cu12==12.6.85
[pip3] nvidia-nvtx-cu12==12.6.77
[pip3] pyzmq==26.4.0
[pip3] sentence-transformers==4.1.0
[pip3] torch==2.7.0
[pip3] torchaudio==2.7.0
[pip3] torchvision==0.22.0
[pip3] transformers==4.52.3
[pip3] transformers-stream-generator==0.0.5
[pip3] triton==3.3.0
[pip3] tritonclient==2.51.0
[pip3] vector-quantize-pytorch==1.21.2
[conda] Could not collect
==============================
vLLM Info
==============================
ROCM Version : Could not collect
Neuron SDK Version : N/A
vLLM Version : 0.8.5.dev373+g947f2f537.d20250430 (git sha: 947f2f537, date: 20250430)
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0 GPU1 GPU2 GPU3 NIC0 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X NV12 NV12 NV12 PIX 0-39 0 N/A
GPU1 NV12 X NV12 NV12 PIX 0-39 0 N/A
GPU2 NV12 NV12 X NV12 NODE 0-39 0 N/A
GPU3 NV12 NV12 NV12 X NODE 0-39 0 N/A
NIC0 PIX PIX NODE NODE X
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
NIC Legend:
NIC0: mlx5_0
==============================
Environment Variables
==============================
NVIDIA_VISIBLE_DEVICES=GPU-ab7b2426-46e6-acb5-8a4f-c66f1bcdd634,GPU-7a9ed403-35ee-85d4-7f69-ad94befd93ac,GPU-657cf630-1798-0bcb-d58d-8ffc43dbf766,GPU-9ccfb583-6b42-463b-ecbd-4cea0695f518
VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
VLLM_WORKER_MULTIPROC_METHOD=fork
VLLM_USAGE_SOURCE=production-docker-image
TORCH_NCCL_HEARTBEAT_TIMEOUT_SEC=15
CUDA_VISIBLE_DEVICES=0,1,2,3
CUDA_VISIBLE_DEVICES=0,1,2,3
TORCH_NCCL_DUMP_ON_TIMEOUT=0
LD_LIBRARY_PATH=/opt/vllm/lib/python3.12/site-packages/nvidia/nvtx/lib:/opt/vllm/lib/python3.12/site-packages/nvidia/cuda_runtime/lib:/opt/vllm/lib/python3.12/site-packages/nvidia/cuda_nvrtc/lib:
VLLM_NO_USAGE_STATS=1
VLLM_USE_V1=0
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY
🐛 Describe the bug
The test_gpu_utilization.py test is failing in CI but I can also reproduce it on a 80GB A100 GPU.
In https://github.com/vllm-project/vllm/blob/main/tests/entrypoints/llm/test_gpu_utilization.py, 3 LLM instances are started at the same time with 0.3 memory utilization. The first one is able to start but for the second one, determine_available_memory() returns a negative number. This is because the code below seems to fail to take into account that non-torch allocated memory could be from other processes. So although the memory is still 70% free when the second instance comes up, it think that it is out of memory.
non_torch_allocations = total_allocated_bytes - torch_allocated_bytes
if non_torch_allocations > 0:
peak_memory += non_torch_allocations
available_kv_cache_memory = (
total_gpu_memory * self.cache_config.gpu_memory_utilization -
peak_memory)
Before submitting a new issue...
- Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.