Replies: 1 comment
-
|
#3554 May be relavant |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Why PD disable radix cache and page size =1 get better performance for nixl backend?
for radix cache, prefix cache should reduce prefill compute, but results in performance drop.
for page size, I think large page size should be helpful for transfer large block of data, but results in performance drop.
Beta Was this translation helpful? Give feedback.
All reactions