upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-31 15:10:59 -05:00
Code Issues Packages Projects Releases Wiki Activity
Files
sl/dio-test
llama.cpp/examples/server/server.cpp
Pavel Fatin 1b17ed7ab6 Direct I/O and Transparent HugePages 
--direct-io for bypassing page cache (and using THP on Linux)

Up to 3-6x faster uncached loading, fewer pageouts, no page cache pollution.
2024-05-21 01:35:23 +02:00

154 KiB
Raw Permalink Blame History

View Raw
Reference in New Issue View Git Blame Copy Permalink