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

[CI] Add launch modes and available blocks tests in e2e test #57

Merged
merged 4 commits into from
Oct 17, 2024
Merged

Conversation

s5u13b
Copy link
Contributor

@s5u13b s5u13b commented Oct 16, 2024

  1. Add launch modes test to ensure that llumnix can launch correctly under different '--disable-init-instance-by-manager' and '--disable-fixed-node-init-instance' combinations.
  2. Add available blocks test to ensure that migrations do not cause kv cache blocks leaking.

tests/e2e_test/test_e2e.py Outdated Show resolved Hide resolved
tests/e2e_test/test_migration.py Outdated Show resolved Hide resolved
Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 36931.39 100444.21 187704.16 219351.37 232893.84 109926.42
decode p25 p50 p75 p95 p99 mean
latency(ms) 52.32 57.39 70.24 114.02 194.76 66.69

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 176.00 MB 192.00 MB 312.00 MB 352.00 MB 472.00 MB 544.00 MB 560.00 MB 592.00 MB 688.00 MB 960.00 MB
rpc_speed(GB/s) 1.03 1.51 1.73 1.87 1.97 2.08 2.11 2.14 2.20 2.26 2.23 2.28 2.29 2.34 2.37 2.36 2.38 2.44 2.30 2.44 2.46 2.47 2.68 2.74 3.00 2.54 3.16 2.99 3.28 3.38
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 152.00 MB 160.00 MB 168.00 MB 176.00 MB 200.00 MB 312.00 MB 352.00 MB 472.00 MB 560.00 MB 688.00 MB
gloo_speed(GB/s) 0.92 1.64 2.04 2.27 2.37 2.76 2.89 2.30 3.01 3.11 2.69 2.73 3.12 2.62 3.85 2.82 2.67 2.60 2.27 1.76 3.10 2.90 2.46 2.64 2.72 2.37 2.58
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 280.00 MB 312.00 MB 320.00 MB 328.00 MB 472.00 MB 544.00 MB
nccl_speed(GB/s) 0.19 0.45 0.73 0.95 1.09 1.37 1.17 1.39 1.69 1.66 1.78 2.02 2.15 2.70 2.49 2.91 3.00 2.25 3.16 2.91 3.54 2.93 3.20 4.06 1.37 4.46 4.39 4.07 6.40 4.80

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 200.00 MB 312.00 MB 320.00 MB 328.00 MB 472.00 MB 552.00 MB 560.00 MB 576.00 MB
rpc_speed(GB/s) 1.03 1.51 1.79 1.90 1.98 2.08 2.15 2.13 2.20 2.24 2.34 2.36 2.26 2.33 2.24 2.39 2.38 2.19 2.44 2.43 2.57 2.22 2.44 2.80 2.73 2.56 3.12 3.18 3.21 2.89
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 312.00 MB 328.00 MB 472.00 MB 552.00 MB 560.00 MB 592.00 MB 736.00 MB 960.00 MB
gloo_speed(GB/s) 1.09 1.64 2.00 2.38 2.63 2.84 2.95 3.12 2.98 3.02 3.16 3.51 2.89 3.59 2.85 2.46 2.95 2.84 2.99 2.58 0.84 3.15 2.93 2.43 3.02 3.10 3.21 2.02 3.19 3.05 2.98
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 152.00 MB 160.00 MB 168.00 MB 176.00 MB 192.00 MB 312.00 MB 320.00 MB 376.00 MB 472.00 MB 536.00 MB 544.00 MB 560.00 MB 576.00 MB
nccl_speed(GB/s) 0.19 0.52 0.56 0.87 1.18 1.27 1.32 1.41 1.75 2.00 1.83 1.91 2.00 2.19 2.68 2.62 3.01 2.91 2.52 3.32 4.00 0.74 4.19 3.64 5.08 4.87 4.84 3.85 3.96 4.38

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 25919.14 102694.89 205764.14 243420.55 250292.08 110472.42
decode p25 p50 p75 p95 p99 mean
latency(ms) 52.85 57.82 69.51 130.87 322.97 70.18

@s5u13b s5u13b merged commit 8d32ee9 into main Oct 17, 2024
14 checks passed
@s5u13b s5u13b deleted the refine-ci branch October 17, 2024 10:32
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.

3 participants