Kafka broker load on Linux processing latency
Test the impact of I/O pressure on Kafka broker throughput and partition synchronization on Linux hosts. As disk I/O usage increases on the broker's host, pipeline performance will decrease, but should stay within expected bounds.
IO
Linux
4 minutes
How this Scenario works
This Scenario generates I/O load on the Linux host where your Kafka broker runs, simulating conditions like disk contention from co-located workloads or degraded storage performance. It lets you observe how your Kafka cluster handles constrained I/O at the host level.
Why run this Scenario?
- Validate that Kafka broker performance degrades gracefully under I/O pressure on bare-metal or VM-based Linux infrastructure.
- Identify I/O throughput thresholds where partition replication lag becomes a risk.
- Verify that host-level monitoring detects I/O saturation and triggers alerts before pipeline throughput is impacted.
- Understand the relationship between Linux host I/O capacity and Kafka cluster health.
What to expect when you run it
When disk I/O load increases on a Linux host running a Kafka broker, pipeline throughput decreases predictably and partition replication remains synchronized within acceptable thresholds.
More Scenarios to try
Avoid downtime. Use Gremlin to turn failure into resilience.
Gremlin empowers you to proactively root out failure before it causes downtime. See how you can harness chaos to build resilient systems by requesting a demo of Gremlin.
.svg)