xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-24 16:46:13 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Metrics] Fix KV cache usage percent metric multiproc (#28792)

Browse Source 
The `vllm:kv_cache_usage_perc` Gauge metric is missing `multiprocess_mode="mostrecent"` and ends up returning

```
vllm:kv_cache_usage_perc{engine="0",model_name="Qwen/Qwen3-VL-8B-Instruct",pid="277"} 0.0
vllm:kv_cache_usage_perc{engine="0",model_name="Qwen/Qwen3-VL-8B-Instruct",pid="275"} 0.0
vllm:kv_cache_usage_perc{engine="0",model_name="Qwen/Qwen3-VL-8B-Instruct",pid="273"} 0.6530455880475035
...
```

The deprecated `vllm:gpu_cache_usage_perc` Gauge metric has `multiprocess_mode="mostrecent"`.

Signed-off-by: Jae-Won Chung <jwnchung@umich.edu>
This commit is contained in:
Jae-Won Chung 2025-11-17 04:54:15 -05:00 committed by GitHub
parent ab01cd14e5
commit d4acf518d0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
vllm/v1/metrics/loggers.py
View File

@ -494,6 +494,7 @@ class PrometheusStatLogger(AggregateStatLoggerBase):
gauge_kv_cache_usage = self._gauge_cls(
name="vllm:kv_cache_usage_perc",
documentation="KV-cache usage. 1 means 100 percent usage.",
multiprocess_mode="mostrecent",
labelnames=labelnames,
)
self.gauge_kv_cache_usage = make_per_engine(