Building Data Streaming Applications with Apache Kafka
by Chanchal Singh, Manisha Sethi, Manish Kumar, Anshul Joshi
Hard drives
You should try to estimate the amount of hard drive space required per broker along with the number of disk drives per broker. Multiple drives help in achieving good throughput. You should also not share Kafka data drives with Kafka logs, Zookeeper data, or other OS file system data. This ensures good latency.
Let’s first talk about how to determine disk space requirements per broker. You should first estimate your average message size. You should also estimate your average message throughput and for how many days you would like to keep the messages in the Kafka cluster.
Based on these estimates, you can calculate the space per broker using the following formula:
(Message Size * Write Throughput * Message Retention Period * Replication ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access