Zihao Ye (@ye_combinator) 's Twitter Profile
Zihao Ye

@ye_combinator

Building flashinfer (github.com/flashinfer-ai/…)

ID: 916605919210827777

linkhttps://homes.cs.washington.edu/~zhye/ calendar_today07-10-2017 10:07:35

118 Tweet

1,1K Followers

511 Following

Zihao Ye (@ye_combinator) 's Twitter Profile Photo
Zihao Ye
@ye_combinator

3 months ago

Check out the intra-kernel profiler in flashinfer to visualize the timeline of each SM/warpgroup in the lifecycle of a CUDA persistent kernel: github.com/flashinfer-ai/… You can clearly understand how tensor/cuda cores overlapping, variable length load-balancing and fusion works.

Check out the intra-kernel profiler in flashinfer to visualize the timeline of each SM/warpgroup in the lifecycle of a CUDA persistent kernel: github.com/flashinfer-ai/… You can clearly understand how tensor/cuda cores overlapping, variable length load-balancing and fusion works.