Usage Stats Collection #2852
Merged
Usage Stats Collection #2852
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[ROCm] Fix build problem resulted from previous commit related to FP8 kv-cache support (vllm-project#2790) Add documentation on how to do incremental builds (vllm-project#2796) [Ray] Integration compiled DAG off by default (vllm-project#2471) Disable custom all reduce by default (vllm-project#2808) add usage context removed usage_context from Engine_args Move IO to another process added http request [ROCm] support Radeon™ 7900 series (gfx1100) without using flash-attention (vllm-project#2768) Add documentation section about LoRA (vllm-project#2834) Refactor 2 awq gemm kernels into m16nXk32 (vllm-project#2723) Co-authored-by: Chunan Zeng <chunanzeng@Chunans-Air.attlocal.net> Added additional arg for from_engine_args comments
mgoin
reviewed
Feb 20, 2024
vllm/usage/usage_lib.py
Outdated
Comment on lines
14
to
16
| _USAGE_STATS_FILE = os.path.join( | ||
| os.path.dirname(os.path.abspath(__file__)), | ||
| 'usage_stats.json') #File path to store usage data locally |
There was a problem hiding this comment.
Since the package can be installed in many different ways/places, it might make more sense to have one clear path such as in the user's directory for documentation. Maybe ~/.config/vllm/usage_stats.json or ~/.cache/vllm/usage_stats.json?
There was a problem hiding this comment.
Ah yes! @yhu422 please do ~/.config/vllm/usage_stats.json and respect XDG_CONFIG_HOME env var.
vllm/usage/usage_lib.py
Outdated
| self.vllm_version = pkg_resources.get_distribution("vllm").version | ||
| self.model = model | ||
| self.log_time = _get_current_timestamp_ns() | ||
| self.num_cpu = os.cpu_count() |
There was a problem hiding this comment.
It would be good to get the type of CPU as well, such as it's product name so you can be aware of what ISA extensions are available for performance
simon-mo
reviewed
Feb 26, 2024
vllm/usage/usage_lib.py
Outdated
Comment on lines
14
to
16
| _USAGE_STATS_FILE = os.path.join( | ||
| os.path.dirname(os.path.abspath(__file__)), | ||
| 'usage_stats.json') #File path to store usage data locally |
There was a problem hiding this comment.
Ah yes! @yhu422 please do ~/.config/vllm/usage_stats.json and respect XDG_CONFIG_HOME env var.
vllm/usage/usage_lib.py
Outdated
| self.vllm_version = pkg_resources.get_distribution("vllm").version | ||
| self.model = model | ||
| self.log_time = _get_current_timestamp_ns() | ||
| self.num_cpu = os.cpu_count() |
|
I made a pass. The current output is as follows {
"provider": "GCP",
"num_cpu": 24,
"cpu_type": "Intel(R) Xeon(R) CPU @ 2.20GHz",
"cpu_family_model_stepping": "6,85,7",
"total_memory": 101261135872,
"architecture": "x86_64",
"platform": "Linux-5.10.0-28-cloud-amd64-x86_64-with-glibc2.31",
"gpu_count": 2,
"gpu_type": "NVIDIA L4",
"gpu_memory_per_device": 23580639232,
"model_architecture": "LlamaForCausalLM",
"vllm_version": "0.3.3+cu123",
"context": "OPENAI_API_SERVER",
"log_time": 1710635056116821000,
"source": "production",
"dtype": "torch.float16",
"tensor_parallel_size": 2,
"block_size": 16,
"gpu_memory_utilization": 0.9,
"quantization": null,
"kv_cache_dtype": "auto",
"enable_lora": false,
"enable_prefix_caching": false,
"enforce_eager": false,
"disable_custom_all_reduce": true
}There are three remaining items before merge (to be completed by @simon-mo)
|
simon-mo
added
the
release-blocker
simon-mo
approved these changes
Mar 29, 2024
xjpang
pushed a commit
to xjpang/vllm
that referenced
this pull request
Mar 31, 2024
Temirulan
pushed a commit
to Temirulan/vllm-whisper
that referenced
this pull request
Sep 6, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Issue: #1376
Added usage_lib.py: Collect information such as GPU type, vllm version, cloud provider upon engine initialization. The info is written to a local file and sent to server. The collection can be disabled through setting environment variables or config file.
Added arguments to from_engine_args(...) to pass in entry point
TODO: Deploy a vector.dev http server to receive usage info.