Python Profiling and Performance Tuning

• 성능 측정과 개선
• 2016-08-14 (일요일) 14:40 - 15:20
• 영어
• 103
• Joe Gordon / jogo
• Photography and recording is allowed

Video

PDF

Description

Pinterest decreased latency and shrunk their front-end python based fleet by over 40% with less than 100 lines of python. This was done by understanding how a significant amount of latency and consumed resources were the result of gevent’s cooperative multitasking. 3 different python profilers, including one running on production traffic, were then used together to identify and fix bottlenecks and inefficiencies in the code base. This talk will cover how Pinterest shrunk their front-end fleet, the techniques and tooling used to improve our performance and how they can be applied elsewhere.

Comments