Red Hat Service Interconnect (Skupper) sizing guide

Issue

How does sizing affect performance for a Skupper (Red Hat Service Interconnect) site?
What recommendations are available for sizing a Skupper (Red Hat Service Interconnect) site?

Resolution

See the sizing guide PDF.

The benchmark results in the PDF aren’t the last word. They depend on the specifics of our test environment. To get a better idea of how Skupper performs in your environment, you can run these benchmarks yourself.

Router CPU

The primary factor to consider when scaling Skupper for your workload is router CPU. (Note that due to the nature of cluster ingress and connection routing, it’s important to focus on scaling the router vertically, not horizontally.)

Two CPU cores (2,000 millicores) per router is a good starting point. It includes some headroom and provides low latencies for a large set of workloads.

If the peak throughput required by your workload is low, it is possible to achieve satisfactory latencies with less router CPU.

Some workloads are sensitive to network latency. In these cases, the overhead introduced by the router can limit the achievable throughput. This is when CPU amounts higher than two cores per router may be required.

On the flip side, some workloads are tolerant of network latency. In these cases, one core or less may be sufficient.

Router memory

Router memory use scales with the number of open connections. In general, a good starting point is 4G.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.