Skip to content

Instantly share code, notes, and snippets.

@noteed

noteed/README.md

Last active July 22, 2020 19:16
Show Gist options
  • Save noteed/7c5bea8b8c3dd4a1d20783aa8bf4c583 to your computer and use it in GitHub Desktop.
Save noteed/7c5bea8b8c3dd4a1d20783aa8bf4c583 to your computer and use it in GitHub Desktop.
Download ZIP
A sense of speed
Raw
README.md

A sense of speed

See: https://computers-are-fast.github.io/

Reading data from disk

Create a file system in RAM:

$ mkdir ramdir
$ mount -t tmpfs -o size=500m tmpfs ramdir

dd if=debian-10.1.0-amd64-virtualbox.box bs=1MB of=ramdir/a
1056+1 records in
1056+1 records out
1056625836 bytes (1.1 GB, 1008 MiB) copied, 0.408706 s, 2.6 GB/s

dd if=debian-10.1.0-amd64-virtualbox.box bs=1MB > /dev/null
1056+1 records in
1056+1 records out
1056625836 bytes (1.1 GB, 1008 MiB) copied, 0.157488 s, 6.7 GB/s

Generating data

At the time of writing, the latest commit of coreutils' seq.c is https://github.com/coreutils/coreutils/commit/1c8050cc4d862a905ba4e91ac6df2f3c1e501649

It contains a comment that shows that one way to determine the throughput of seq is

$ seq 0 200 inf | pv > /dev/null
 171MiB 0:00:08 [21.6MiB/s] [         <=>                                                 ]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment