Hello Kyle,
I was exploring few options to use Jepsen as Benchmarking tool. I could generate a mix workload but when we talk about performance/benchmark we try to give deterministic workload (Eg: 20% read, 80% write or 100% put request, etc) and a better way to map throughput and latency.
Please suggest if there is anyway to do that.
Thanks,
Sapan