bsdinis / echo

Example echo servers for performance comparison

Echo Server comparison

Comparing different PLs/frameworks as it pertains to the performance of simple echo servers.

API

Server

The server should support the following CLI options:

• [hostname]: hostname to listen on
• [port]: port to listen on
• -j, --n-cores: integer, number of cores to use (default: number of cores in the machine)
• -b, --backlog: integer, backlog of the listening socket

Client

The client should support the following CLI options:

• [hostname]: hostname to connect to
• [port]: port to connect to
• -j, --n-cores: integer, number of cores to use for concurrency (default: number of cores in the machine)
• -d, --duration: duration of the experiment
• -w, --warmup: duration of the warmup cycle
• -s, --start: start instatnt
• -m, --message-size: size of the message to send (parseable, like 1MB or 256KiB)

Output

The clients SHALL output a list of latencies in microseconds. There MUST be at least one line with Messagte Size: Z, in bytes. In the event there are multiple such lines, they should be identical. Each client will output a Start: <ID> A.B and an End: <ID> X.Y, such that X.Y - A.B will give the elapsed time in seconds. In the event of multiple Starts and Ends per <ID>, the considered Start will be the minimum value and the considered End the maximum value.