From 8a5c053d391dae7b5f481bb44c1302fb6f39d39d Mon Sep 17 00:00:00 2001 From: aitoehigie Date: Sat, 13 Oct 2018 02:34:58 +0100 Subject: [PATCH] Add a concise benchmarks section to the README.md file --- README.md | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index b565b00..7b13cfa 100644 --- a/README.md +++ b/README.md @@ -105,14 +105,32 @@ api = responder.API(enable_hsts=True) Boom. ✨🍰✨ # Performance +The objective of the benchmark tests is not testing deployment (like uwsgi vs gunicorn and etc) but instead test the performance of python-response against other popular Python web frameworks. + +Methodology +The results below were gotten running the performance tests on a Lenovo W530, Intel(R) Core(TM) i7-3740QM CPU @ 2.70GHz, MEM: 32GB, Linux Mint 19. I used Python 3.7.0 with the WRK utility with params: +wrk -d20s -t10 -c200 (i.e. 10 threads and 200 connections). + +1. Simple "Hello World" benchmark + + python-responder v0.0.1 (Master branch) + Requests/sec: 1368.23 + Transfer/sec: 163.01KB + + Django v2.1.2 (i18n == False) + Requests/sec: 544.54 + Transfer/sec: 103.18KB + + Django v2.1.2 (i18n == True) + Requests/sec: 535.12 + Transfer/sec: 101.38KB + + Flask v1.0.2 + Requests/sec: 896.24 + Transfer/sec: 144.41KB + - python-responder v0.0.1 [stats] - Requests/sec: 952.54 - Transfer/sec: 119.07KB - Django v2.1.2 (i18n == False) [stats] - Requests/sec: 520.87 - Transfer/sec: 98.68KB # The Basic Idea