Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Misc] Usage Doc for Prefill-decoding Disaggregation #71

Merged
merged 6 commits into from
Dec 5, 2024
Merged

Conversation

Xinyi-ECNU
Copy link
Contributor

No description provided.

Copy link

migration_size 8.00 MB 16.00 MB 24.00 MB 32.00 MB 40.00 MB 48.00 MB 56.00 MB 64.00 MB 72.00 MB 80.00 MB 88.00 MB 96.00 MB 104.00 MB 112.00 MB 120.00 MB 128.00 MB 136.00 MB 144.00 MB 152.00 MB 160.00 MB 168.00 MB 176.00 MB 184.00 MB 192.00 MB 200.00 MB 208.00 MB 216.00 MB 224.00 MB 248.00 MB 352.00 MB
rpc_speed(GB/s) 1.05 1.58 1.88 2.00 2.16 2.23 2.25 2.25 2.39 2.48 2.45 2.43 2.56 2.59 2.55 2.64 2.68 2.54 2.61 2.67 2.41 2.79 2.62 2.64 2.52 2.47 2.75 2.62 2.86 3.24
migration_size 8.00 MB 16.00 MB 24.00 MB 32.00 MB 40.00 MB 48.00 MB 56.00 MB 64.00 MB 72.00 MB 80.00 MB 88.00 MB 96.00 MB 104.00 MB 112.00 MB 120.00 MB 128.00 MB 144.00 MB 152.00 MB 160.00 MB 168.00 MB 176.00 MB 192.00 MB 240.00 MB 280.00 MB 320.00 MB
gloo_speed(GB/s) 1.03 1.67 2.01 2.33 2.45 2.75 2.82 2.87 3.20 3.17 3.28 2.99 3.27 3.36 3.59 3.09 2.80 2.44 2.99 3.48 3.25 2.33 2.47 3.52 3.68

docs/Prefill-decoding_Disaggregation.md Show resolved Hide resolved
docs/Prefill-decoding_Disaggregation.md Outdated Show resolved Hide resolved
docs/Prefill-decoding_Disaggregation.md Outdated Show resolved Hide resolved
@KuilongCui
Copy link
Contributor

KuilongCui commented Nov 13, 2024

as there are some bugs for migraiton, the ci didn't pass.
I will fix it later.
Now, just re-run the failed test

@KuilongCui
Copy link
Contributor

Missing a description of the failover mechanism.

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 13478.72 51380.54 113524.15 136427.30 150863.80 64445.81
decode p25 p50 p75 p95 p99 mean
latency(ms) 51.51 55.28 64.16 91.39 232.25 61.59

@CLAassistant
Copy link

CLAassistant commented Nov 14, 2024

CLA assistant check
All committers have signed the CLA.

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 22331.82 69087.93 101327.92 132419.66 143763.24 63773.35
decode p25 p50 p75 p95 p99 mean
latency(ms) 52.22 56.03 65.43 108.66 272.55 64.03

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 19010.35 51957.98 109831.90 146296.89 178691.99 61143.34
decode p25 p50 p75 p95 p99 mean
latency(ms) 53.12 58.57 76.49 119.23 264.28 69.83

Copy link

github-actions bot commented Dec 4, 2024

prefill p25 p50 p75 p95 p99 mean
latency(ms) 19210.32 63736.45 112487.70 145522.58 172127.51 65107.77
decode p25 p50 p75 p95 p99 mean
latency(ms) 52.41 56.56 66.74 118.99 303.16 67.98

@zhypku zhypku requested a review from KuilongCui December 4, 2024 07:46
@KuilongCui
Copy link
Contributor

LGTM :)

Copy link

github-actions bot commented Dec 5, 2024

prefill p25 p50 p75 p95 p99 mean
latency(ms) 18272.04 64507.25 106716.45 135297.28 146702.78 65166.56
decode p25 p50 p75 p95 p99 mean
latency(ms) 51.56 56.41 67.26 128.60 338.44 66.81

@zhypku zhypku merged commit ba15665 into main Dec 5, 2024
13 of 14 checks passed
@zhypku zhypku deleted the pdd_doc branch December 5, 2024 10:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants