mirrors/granian
1
0
Fork 0
mirror of https://github.com/emmett-framework/granian.git synced 2025-12-23 05:36:49 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
granian/benchmarks/vs.md
Giovanni Barillari 02f683950c
Update benchmarks
2025-05-21 14:29:21 +02:00

4.5 KiB
Raw Permalink Blame History

Granian benchmarks

VS 3rd party comparison

Run at: Wed 21 May 2025, 00:21
Environment: AMD Ryzen 7 5700X @ Ubuntu 24.04 (CPUs: 16)
Python version: 3.12
Granian version: 2.3.1

Note: unless otherwise specified, all benchmarks are run with 1 server worker and 1 thread.

ASGI

Server Total requests RPS avg latency max latency
Granian Asgi get 10KB (c128) 1209492 121065 1.056ms 21.722ms
Granian Asgi echo 10KB (iter) (c256) 604180 60601 4.218ms 33.901ms
Uvicorn H11 get 10KB (c64) 95321 9532 6.707ms 16.222ms
Uvicorn H11 echo 10KB (iter) (c64) 83145 8315 7.682ms 22.916ms
Uvicorn Httptools get 10KB (c128) 396594 39694 3.22ms 16.505ms
Uvicorn Httptools echo 10KB (iter) (c128) 349481 34974 3.652ms 17.761ms
Hypercorn get 10KB (c128) 65706 6575 19.43ms 192.337ms
Hypercorn echo 10KB (iter) (c128) 57788 5784 22.07ms 187.016ms

WSGI

Server Total requests RPS avg latency max latency
Granian Wsgi get 10KB (c64) 1234817 123487 0.517ms 3.097ms
Granian Wsgi echo 10KB (iter) (c64) 866084 86606 0.738ms 3.67ms
Gunicorn Gthread get 10KB (c64) 63449 6345 10.068ms 13.108ms
Gunicorn Gthread echo 10KB (iter) (c64) 44999 4500 14.192ms 25.237ms
Gunicorn Gevent get 10KB (c64) 94102 9411 3.541ms 7460.271ms
Gunicorn Gevent echo 10KB (iter) (c64) 63017 6302 4.813ms 7860.646ms
Uwsgi get 10KB (c64) 165575 16558 3.858ms 5.763ms
Uwsgi echo 10KB (iter) (c256) 193382 19386 10.264ms 6588.85ms

HTTP/2

Server Total requests RPS avg latency max latency
Granian Asgi get 10KB (c64) 670752 67079 0.952ms 4.25ms
Granian Asgi echo 10KB (iter) (c512) 122105 12274 41.544ms 157.271ms
Hypercorn get 10KB (c128) 41415 4143 30.809ms 399.949ms
Hypercorn echo 10KB (iter) (c128) 31932 3195 39.934ms 82.546ms

ASGI file responses

Server Total requests RPS avg latency max latency
Granian (pathsend) (c128) 500730 50116 2.55ms 30.838ms
Uvicorn H11 (c128) 95279 9533 13.408ms 317.559ms
Uvicorn Httptools (c64) 151960 15196 4.204ms 10.432ms
Hypercorn (c128) 65378 6542 19.515ms 30.242ms

Long I/O

Plain text response comparison simulating long I/O waits (10ms and 100ms).

Server Total requests RPS avg latency max latency
Granian Rsgi 10ms (c512) 469509 47111 10.849ms 79.164ms
Granian Rsgi 100ms (c512) 50291 5054 100.853ms 104.491ms
Granian Asgi 10ms (c512) 463502 46495 10.99ms 61.469ms
Granian Asgi 100ms (c512) 50284 5050 101.0ms 104.911ms
Granian Wsgi 10ms (c512) 418746 42024 12.165ms 47.537ms
Granian Wsgi 100ms (c512) 50294 5051 100.391ms 104.889ms
Uvicorn Httptools 10ms (c512) 310265 31111 16.425ms 153.543ms
Uvicorn Httptools 100ms (c512) 50276 5046 100.763ms 108.566ms
Hypercorn 10ms (c128) 64491 6456 19.781ms 194.227ms
Hypercorn 100ms (c128) 64360 6441 19.829ms 190.223ms
Gunicorn Gevent 10ms (c128) 85902 8598 14.86ms 26.707ms
Gunicorn Gevent 100ms (c512) 50288 5052 100.758ms 138.529ms
Uwsgi 10ms (c512) 163394 16412 20.75ms 7610.089ms
Uwsgi 100ms (c128) 164639 16474 7.696ms 3354.149ms

Websockets

Websocket broadcasting comparison with concurrent clients sending a predefined amount of messages and receiving those messages from all the connected clients. The benchmark takes the time required for the test to run and compute the relevant throughput (in messages per second).

Clients Server Send throughput Receive throughput Combined throughput
8 Granian Rsgi 1022244 213279 239939
8 Granian Asgi 1029782 216303 243341
8 Uvicorn H11 610660 125721 141436
8 Hypercorn 580836 90975 102347
16 Granian Rsgi 1937499 226845 241023
16 Granian Asgi 1935757 226807 240983
16 Uvicorn H11 1261027 128309 136329
16 Hypercorn 1248253 98296 104439
32 Granian Rsgi 3572750 225997 233059
32 Granian Asgi 3597453 225285 232325
32 Uvicorn H11 N/A N/A N/A
32 Hypercorn N/A N/A N/A