/plushcap/analysis/aiven/benchmarking-kafka-write-throughput-performance-2022-update

Benchmarking Aiven for Apache Kafka® throughput [2022 UPDATE]

What's this blog post about?

Aiven has recently measured the throughput performance of Apache Kafka on its platform, updating benchmarking results from 2017 and 2019. The tests were conducted on three standard plans - Business-4, Business-8, and Premium-6x-8 - across three major cloud providers: AWS, GCP, and Azure. The goal was to determine the best plan for various application workloads. The benchmarking results showed that the maximum sustainable write throughput varied depending on the cloud provider and the Aiven plan used. For example, in the Business-4 plan, the maximum bytes in rate from clients were 37.0 MB/s (AWS), 30.5 MB/s (GCP), and 32.7 MB/s (Azure). In contrast, for the Premium-6x-8 plan, the respective figures increased to 123 MB/s, 147 MB/s, and 137 MB/s. The cost per MB/s per month remained relatively consistent across different plans, with only a slight increase as the plan's capacity was upgraded. Aiven for Apache Kafka is designed to ensure that it does not hit IOPS limits in any 24-hour period, making it suitable for production workloads.

Company
Aiven

Date published
Nov. 10, 2022

Author(s)
Mikhail Kokho

Word count
850

Hacker News points
None found.

Language
English


By Matt Makai. 2021-2024.