3 ms·
10-100k RPS is throughput, not latency. There is a well-known latency-throughput tradeoff. For example, Erlang’s BEAM VM is optimized for low latency, while JVM
by sesm 3y ago
10-100k RPS is throughput, not latency. There is a well-known latency-throughput tradeoff. For example, Erlang’s BEAM VM is optimized for low latency, while JVM is optimized for high throughput.