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

[Core] Support for Scheduling-defined Prefill-Decode Disaggregation feature #15

Merged
merged 13 commits into from
Oct 14, 2024

Conversation

Xinyi-ECNU
Copy link
Contributor

@Xinyi-ECNU Xinyi-ECNU commented Aug 23, 2024

Design for introducing cluster-level prefill-decode disaggregation design to Llumnix. Based on dynamic rescheduling of requests in Llumnix, this design allows Llumnix to manage prefill/decoding instances and the scheduling of requests on these instances. Specifically, this PR designs broader scheduling semantics, enabling the rules for PD disaggregation to be expressed as customized policies within Llumnix.

llumnix/arg_utils.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/llm_engine.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/llm_engine.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/llm_engine.py Outdated Show resolved Hide resolved
@Xinyi-ECNU Xinyi-ECNU requested a review from zhypku August 26, 2024 06:02
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/global_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/llumlet/local_migration_scheduler.py Outdated Show resolved Hide resolved
Copy link
Collaborator

@zhypku zhypku left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'd like to call this feature as: scheduling-defined pdd :)

llumnix/arg_utils.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/llm_engine.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/llm_engine.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/global_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/global_scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/config.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/llm_engine_manager.py Outdated Show resolved Hide resolved
llumnix/llm_engine_manager.py Outdated Show resolved Hide resolved
llumnix/llm_engine_manager.py Outdated Show resolved Hide resolved
llumnix/llm_engine_manager.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
@Xinyi-ECNU Xinyi-ECNU changed the title [Core] Support for Prefill-Decode Disaggregation feature [Core] Support for Scheduling-defined Prefill-Decode Disaggregation feature Aug 27, 2024
@CLAassistant
Copy link

CLAassistant commented Aug 28, 2024

CLA assistant check
All committers have signed the CLA.

llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Outdated Show resolved Hide resolved
llumnix/config.py Outdated Show resolved Hide resolved
llumnix/config.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/global_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
llumnix/llm_engine_manager.py Outdated Show resolved Hide resolved
llumnix/llumlet/llumlet.py Outdated Show resolved Hide resolved
llumnix/llumlet/migration_coordinator.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/backends/vllm/scheduler.py Outdated Show resolved Hide resolved
llumnix/llumlet/request.py Outdated Show resolved Hide resolved
llumnix/llumlet/llumlet.py Outdated Show resolved Hide resolved
llumnix/backends/backend_interface.py Show resolved Hide resolved
llumnix/llumlet/request.py Outdated Show resolved Hide resolved
llumnix/llumlet/request.py Show resolved Hide resolved
llumnix/llumlet/llumlet.py Show resolved Hide resolved
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 224.00 MB 232.00 MB 280.00 MB 288.00 MB 312.00 MB 344.00 MB 368.00 MB 416.00 MB 424.00 MB 432.00 MB 440.00 MB 496.00 MB 720.00 MB 912.00 MB
rpc_speed(GB/s) 1.05 1.53 1.79 1.95 2.04 2.18 2.18 2.22 2.24 2.34 2.34 2.41 2.43 2.40 2.49 2.51 2.62 2.34 2.34 2.59 2.52 2.50 2.52 2.46 2.64 2.78 2.18 2.49 2.80 2.59 2.76 2.92 3.18 3.05 3.13 3.17 2.93 3.45 3.32 3.06
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 192.00 MB 200.00 MB 240.00 MB 280.00 MB 312.00 MB 384.00 MB 416.00 MB 464.00 MB 480.00 MB 488.00 MB 536.00 MB 544.00 MB 696.00 MB
gloo_speed(GB/s) 0.92 1.60 1.98 2.28 2.44 2.78 2.72 2.99 2.64 2.91 2.90 2.57 2.62 2.68 3.41 2.62 2.36 2.25 2.19 2.13 2.18 2.50 2.92 2.90 3.40 2.56 2.38 2.97 2.51 2.68 2.84 2.63 2.73 2.64 2.81
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 208.00 MB 232.00 MB 240.00 MB 256.00 MB 280.00 MB 312.00 MB 320.00 MB 416.00 MB 424.00 MB 448.00 MB 464.00 MB 488.00 MB 528.00 MB 536.00 MB 752.00 MB
nccl_speed(GB/s) 0.19 0.45 0.68 0.85 1.14 1.33 1.50 1.52 2.00 1.76 1.76 2.24 2.38 2.63 2.26 2.63 2.86 2.58 3.38 3.44 4.23 3.41 3.62 3.31 3.22 4.43 5.22 4.46 2.65 2.94 4.44 5.66 5.35 5.49 3.73 3.54 5.44 3.99 3.96

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 32545.25 99194.50 192927.00 260004.45 292360.40 114263.90
decode p25 p50 p75 p95 p99 mean
latency(ms) 54.14 59.66 73.90 135.57 231.50 72.70

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 216.00 MB 232.00 MB 240.00 MB 248.00 MB 272.00 MB 312.00 MB 352.00 MB 376.00 MB 424.00 MB 464.00 MB 480.00 MB 536.00 MB 544.00 MB 560.00 MB
rpc_speed(GB/s) 1.03 1.52 1.75 1.90 1.99 2.11 2.03 2.13 2.16 2.21 2.28 2.35 2.32 2.36 2.44 2.40 2.42 2.37 2.42 2.39 2.46 2.39 2.64 2.45 2.32 2.61 2.58 2.27 2.63 2.48 2.76 2.72 2.91 2.78 2.93 2.96 3.01 3.08 2.99
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 224.00 MB 264.00 MB 312.00 MB 320.00 MB 416.00 MB 424.00 MB 480.00 MB 536.00 MB
gloo_speed(GB/s) 0.94 1.56 2.08 2.27 2.47 2.53 2.69 2.91 3.05 2.95 2.82 3.05 3.31 3.30 2.61 2.37 2.73 3.03 2.74 2.42 2.36 2.61 3.02 3.19 3.02 2.83 2.57 2.53 2.40 2.85 2.29 2.60 2.61
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 208.00 MB 232.00 MB 280.00 MB 312.00 MB 336.00 MB 416.00 MB 424.00 MB 432.00 MB 472.00 MB 480.00 MB 536.00 MB 560.00 MB 568.00 MB
nccl_speed(GB/s) 0.19 0.45 0.68 0.87 1.04 1.20 1.44 1.67 1.92 1.93 1.96 2.17 2.38 2.23 2.33 2.68 2.90 3.11 2.72 3.08 2.83 4.17 2.83 3.27 5.00 4.47 2.73 3.18 3.45 5.82 6.39 3.52 5.66 7.48 6.22 5.66 4.20

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 28162.75 101213.50 206765.75 237777.95 250164.08 111382.05
decode p25 p50 p75 p95 p99 mean
latency(ms) 53.26 59.09 72.60 130.21 401.89 75.28

llumnix/arg_utils.py Show resolved Hide resolved
llumnix/global_scheduler/migration_scheduler.py Outdated Show resolved Hide resolved
Copy link

github-actions bot commented Oct 8, 2024

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 232.00 MB 280.00 MB 312.00 MB 416.00 MB 424.00 MB 432.00 MB 464.00 MB 472.00 MB 536.00 MB 552.00 MB 912.00 MB
rpc_speed(GB/s) 1.08 1.59 1.86 2.01 2.11 2.18 2.23 2.29 2.30 2.32 2.46 2.43 2.47 2.34 2.63 2.54 2.53 2.49 2.59 2.64 2.71 2.43 2.52 2.51 2.64 2.73 2.43 2.59 2.78 2.88 2.71 3.00 3.11 3.07 3.31 3.12 3.18 3.26 3.73
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 240.00 MB 256.00 MB 264.00 MB 312.00 MB 416.00 MB 464.00 MB 480.00 MB 496.00 MB 560.00 MB 568.00 MB
gloo_speed(GB/s) 0.95 1.55 2.02 2.23 2.31 2.59 2.67 2.72 2.81 3.20 3.21 2.81 3.45 3.11 3.12 2.81 2.83 2.16 1.85 2.66 2.14 2.46 2.20 2.66 2.06 3.30 2.52 3.27 3.69 1.55 2.86 2.35 2.72 2.47 2.67 2.77 2.75 2.83
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 240.00 MB 280.00 MB 312.00 MB 400.00 MB 416.00 MB 424.00 MB 432.00 MB 464.00 MB 480.00 MB 536.00 MB 560.00 MB
nccl_speed(GB/s) 0.18 0.43 0.67 0.99 1.03 1.24 1.43 1.48 1.92 1.81 2.48 2.22 2.59 2.40 2.32 2.59 3.00 3.18 3.41 3.38 3.40 3.40 3.49 3.38 3.49 3.83 2.55 3.04 5.52 5.22 5.93 4.58 4.94 5.32 3.97 3.49

Copy link

github-actions bot commented Oct 8, 2024

prefill p25 p50 p75 p95 p99 mean
latency(ms) 28845.75 126632.50 179699.25 235444.65 245677.13 110102.76
decode p25 p50 p75 p95 p99 mean
latency(ms) 53.66 60.27 74.86 141.30 329.02 74.69

llumnix/arg_utils.py Show resolved Hide resolved
llumnix/arg_utils.py Outdated Show resolved Hide resolved
Copy link

github-actions bot commented Oct 8, 2024

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 224.00 MB 240.00 MB 264.00 MB 280.00 MB 312.00 MB 320.00 MB 416.00 MB 480.00 MB 560.00 MB 704.00 MB
rpc_speed(GB/s) 1.05 1.55 1.78 1.91 2.03 2.06 2.10 2.18 2.25 2.24 2.30 2.35 2.31 2.44 2.41 2.44 2.46 2.51 2.45 2.52 2.55 2.42 2.37 2.64 2.46 2.42 2.69 2.52 2.63 2.82 2.83 2.65 2.99 2.94 3.08 3.32
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 192.00 MB 200.00 MB 232.00 MB 280.00 MB 312.00 MB 320.00 MB 416.00 MB 448.00 MB 456.00 MB 464.00 MB 480.00 MB 536.00 MB 696.00 MB
gloo_speed(GB/s) 0.91 1.55 1.98 2.18 2.22 2.53 2.60 2.78 2.84 2.99 2.94 2.91 2.83 2.47 2.98 2.54 2.82 2.35 2.94 2.30 1.72 2.23 3.34 1.50 2.63 2.60 2.29 2.41 2.70 2.77 2.70 2.47 2.41 2.62 2.84
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 232.00 MB 312.00 MB 320.00 MB 384.00 MB 416.00 MB 424.00 MB 464.00 MB 480.00 MB 536.00 MB 544.00 MB 576.00 MB
nccl_speed(GB/s) 0.19 0.46 0.73 0.98 1.09 1.30 1.63 1.70 1.95 1.78 2.14 2.15 2.98 2.60 2.42 2.67 2.93 3.16 2.95 3.59 3.64 3.52 2.62 3.97 5.17 3.01 4.23 4.01 5.04 5.64 6.17 5.41 5.80 4.02 3.99 4.35

Copy link

github-actions bot commented Oct 8, 2024

prefill p25 p50 p75 p95 p99 mean
latency(ms) 25627.75 106297.00 172871.00 245538.05 281652.55 109409.63
decode p25 p50 p75 p95 p99 mean
latency(ms) 54.52 61.77 80.36 155.31 401.34 78.39

Copy link

github-actions bot commented Oct 8, 2024

prefill p25 p50 p75 p95 p99 mean
latency(ms) 31386.00 79725.00 174082.25 252759.10 307512.08 104876.12
decode p25 p50 p75 p95 p99 mean
latency(ms) 54.12 60.87 77.45 144.30 440.16 79.42

Copy link

github-actions bot commented Oct 8, 2024

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 232.00 MB 240.00 MB 248.00 MB 312.00 MB 320.00 MB 416.00 MB 448.00 MB 464.00 MB 480.00 MB 528.00 MB
rpc_speed(GB/s) 1.05 1.55 1.75 1.95 2.04 2.14 2.16 2.15 2.27 2.24 2.36 2.40 2.36 2.38 2.44 2.46 2.52 2.46 2.52 2.49 2.58 2.45 2.49 2.36 2.45 2.45 2.61 2.50 2.59 2.62 2.78 2.85 2.90 3.12 2.88 2.81 3.21
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 232.00 MB 240.00 MB 272.00 MB 280.00 MB 312.00 MB 416.00 MB 472.00 MB 480.00 MB 536.00 MB 552.00 MB
gloo_speed(GB/s) 0.94 1.59 2.09 2.22 2.47 2.63 2.87 2.90 2.83 2.90 2.78 3.21 3.27 3.20 3.07 3.00 2.76 1.63 2.64 2.70 1.79 2.50 2.77 1.57 2.84 2.87 2.70 2.77 2.36 2.08 2.98 2.29 2.44 2.46 2.47
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 248.00 MB 280.00 MB 312.00 MB 320.00 MB 416.00 MB 424.00 MB 464.00 MB 480.00 MB 536.00 MB 552.00 MB 560.00 MB 912.00 MB
nccl_speed(GB/s) 0.20 0.56 0.73 0.81 1.24 1.29 1.39 1.68 1.80 1.78 1.90 1.92 1.75 2.70 2.92 2.72 2.91 3.44 3.72 3.98 2.29 3.32 3.22 3.68 3.43 3.82 2.62 3.43 4.49 5.75 6.05 5.84 6.35 4.66 4.13 5.22 4.81

llumnix/global_scheduler/global_scheduler.py Outdated Show resolved Hide resolved
llumnix/llumlet/llumlet.py Outdated Show resolved Hide resolved
llumnix/llumlet/llumlet.py Show resolved Hide resolved
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 200.00 MB 312.00 MB 368.00 MB 472.00 MB 560.00 MB 584.00 MB 736.00 MB
rpc_speed(GB/s) 1.04 1.53 1.75 1.92 1.97 2.09 2.13 2.13 2.18 2.26 2.27 2.23 2.38 2.30 2.36 2.43 2.34 2.40 2.58 2.42 2.37 2.42 2.69 2.67 2.75 2.99 3.09 3.28 3.30
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 320.00 MB 336.00 MB 472.00 MB 560.00 MB 688.00 MB
gloo_speed(GB/s) 0.95 1.52 1.93 2.02 2.37 2.41 2.41 2.44 2.55 3.01 3.01 3.53 3.08 3.17 3.56 2.75 1.82 2.67 2.61 2.84 2.36 2.57 2.89 2.37 2.36 2.58 2.50 1.84 2.39
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 288.00 MB 312.00 MB 480.00 MB 536.00 MB 544.00 MB 576.00 MB 696.00 MB 736.00 MB 960.00 MB
nccl_speed(GB/s) 0.19 0.44 0.67 0.90 0.95 1.24 1.24 1.60 1.69 1.89 2.19 1.87 2.36 2.46 2.61 2.53 2.80 2.71 3.57 3.25 3.34 4.82 4.98 3.63 2.60 2.98 5.51 5.19 3.99 4.83 4.42 4.70 4.95 4.68

Copy link

prefill p25 p50 p75 p95 p99 mean
latency(ms) 26800.25 87265.37 157870.81 294178.67 322501.43 111403.61
decode p25 p50 p75 p95 p99 mean
latency(ms) 52.59 57.49 68.09 108.79 165.82 64.83

@Xinyi-ECNU Xinyi-ECNU merged commit ce45945 into main Oct 14, 2024
14 checks passed
@Xinyi-ECNU Xinyi-ECNU deleted the pd_disagg branch October 18, 2024 02:48
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.

6 participants