Catalogue / slack-2017-job-queue

Slack puts Kafka in front of its Redis job queue after Redis ran out of memory

Slack, 2017. Job queue.

Organisation
Slack
Evidence
Reported from production
Source
Scaling Slack's Job Queue

Slack Engineering, published 2017-12-06

Records

Failure#f1

Database contention filled Redis memory, and a full Redis could neither accept nor release jobs

Decision#d1

Added Kafka in front of the Redis job queue instead of replacing Redis

Decision#d2

Enqueue path acknowledges after the Kafka leader only, trading a small risk of lost jobs for latency

Sign in to read what was chosen, what was rejected and what it cost, each with its place in the source. An account is free.