biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Core][Bugfix][Perf] Introduce MQLLMEngine to avoid asyncio OH (#8157)

Browse Source 
Co-authored-by: Nick Hill <nickhill@us.ibm.com>
Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-neuralmagic@users.noreply.github.com>
Co-authored-by: Simon Mo <simon.mo@hey.com>
This commit is contained in:
Alexander Matveev
2024-09-18 09:56:58 -04:00
committed by GitHub
parent 9d104b5beb
commit 7c7714d856
36 changed files with 1464 additions and 1169 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/source/dev/profiling/profiling_index.rst
View File

@@ -21,8 +21,8 @@ Traces can be visualized using https://ui.perfetto.dev/.
.. tip::
To stop the profiler - it flushes out all the profile trace files to the directory. This takes time, for example for about 100 requests worth of data for a llama 70b, it takes about 10 minutes to flush out on a H100.
Set the env variable VLLM_RPC_GET_DATA_TIMEOUT_MS to a big number before you start the server. Say something like 30 minutes.
``export VLLM_RPC_GET_DATA_TIMEOUT_MS=1800000``
Set the env variable VLLM_RPC_TIMEOUT to a big number before you start the server. Say something like 30 minutes.
``export VLLM_RPC_TIMEOUT=1800000``
Example commands and usage:
===========================