Rootconf 2018

On scaling infrastructure and operations


Partition tolerant, highly available redis cluster with haproxy

Submitted by Ayush Goyal (@perfectayush) on Monday, 29 January 2018

Preview video

Section: Full talk Technical level: Intermediate


This talk will present how production engineering team at Helpshift achieved a partition tolerant, highly available redis cluster using redis sentinel and haproxy, without changing a single line of backend code


Following is a brief overview of what will be covered in the talk

  • Overview of Redis Sentinel service for highly availability of redis master.
  • Various approaches/caveats while architecting a partition tolerant, highly available redis cluster.
  • How we achieved high availability of redis cluster using haproxy
  • Various redis, redis sentinel and haproxy config tunings

Speaker bio

I am part of the production engineering team at Helpshift. We are building a remarkable customer service platform using interesting tech and nifty hacks. My job requires me to design and automate production systems for reliability and performance. You can read more about me at



Preview video


Login with Twitter or Google to leave a comment