kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Fix typo in perf_best_practices.md (#857)

Browse Source 
Fix typo.
This commit is contained in:
BasicCoder 2024-02-01 09:58:39 +08:00 committed by GitHub
parent e06f537e08
commit b310ec6751
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/perf_best_practices.md
View File

@ -222,7 +222,7 @@ inputs and outputs.
There currently are two batch scheduler policies: `MAX_UTILIZATION` and
`GUARANTEED_NO_EVICT`.
As explained in the [GPT Manager Ddesign](batch_manager.md#gptmanager-design)
As explained in the [GPT Manager Design](batch_manager.md#gptmanager-design)
section, the scheduling policy can be set to `MAX_UTILIZATION` to pack as many
requests as possible at each iteration of the forward loop, when in-flight
sequence batching is enabled. It maximizes the utilization of the GPUs by