Home/ Questions/Q 4002098
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-20T07:59:05+00:00

This question is specific to Erlang, but may have general implications to other IO

  • 0

This question is specific to Erlang, but may have general implications to other IO libraries. In Erlang, I can write io:format(IOF, "data: ~p", [Data]) and it will print the contents of the variable Data to the file IOF points to. My question is this: is it better, efficiency-wise, to prefer many small io:format() calls, or one huge one? For example, what would be faster? For purposes of demonstration, assume the size of my Data variables is 1KB each in size, and N > 100,000

Scenario A:

io:format(IOF, "data1: ~p", [Data1])
io:format(IOF, "data2: ~p", [Data2])
...
io:format(IOF, "dataN: ~p", [DataN])

Scenario B:

io:format(IOF, "data1: ~p data2: ~p ... dataN: ~p", [Data1, Data2, ..., DataN])

If this difference is negligible I’m not really interested, but if there could be a large difference then I’m interested in knowing why.

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    I’d recommend you to research on file:open/2 modes, one of which is {delayed_write, Size, Delay}. It allows you to buffer data and effectively write it to the device.

    This way either calls wouldn’t probably make a huge difference.

    • 0