We are defining a set of benchmarks for chat protocols to be measured by the Vac-DST team. For two reasons:

• to quantify improvements tackled by Waku Chat team using the FURPS framework: Waku FURPS
• to perform non-regression performance testing for the chat protocols implementation in status-go, for Status app releases

The order of scenarios are in order of priority.

The scenarios are kept minimal, with a limited scope, on purpose. Once this is delivered, we can review what improvements and changes we are making to chat protocols, and when doing FURPS, see if we need more baseline benchmarks to measure said FURPS.

Metrics

Resource Usage

Scenarios should measure the following parameters:

• Bandwidth usage
• CPU usage
• Memory usage
• Other parameters, as specified in scenarios

Vac-DST’s methodology should be applied as usual, providing data aggregation: average, min, max, etc

Waku Metrics

There are a number of Waku-level parameters that COULD be considered as part of those benchmarks:

• Number of connected peers
• Number of peers in gossipsub mesh
• Number of discovered peers
• Number of ongoing filter connections