[systemd-devel] [PATCH 2/2] compress: add benchmark-style test

[Zbigniew Jędrzejewski-Szmek]

On Mon, Jul 07, 2014 at 09:47:15PM +0200, Kay Sievers wrote:
> On Mon, Jul 7, 2014 at 9:40 PM, Zbigniew Jędrzejewski-Szmek
> <zbyszek at in.waw.pl> wrote:
> > On Mon, Jul 07, 2014 at 09:35:06PM +0300, Uoti Urpala wrote:
> >> On Sat, 2014-07-05 at 20:56 +0200, Zbigniew Jędrzejewski-Szmek wrote:
> >> > This is useful to test the behaviour of the compressor for various buffer
> >> > sizes.
> >> >
> >> > Time is limited to a minute per compression, since otherwise, when LZ4
> >> > takes more than a second which is necessary to reduce the noise, XZ
> >> > takes more than 10 minutes.
> >> >
> >> > % build/test-compress-benchmark (without time limit)
> >> > XZ: compressed & decompressed 2535300963 bytes in 794.57s (3.04MiB/s), mean compresion 99.95%, skipped 3570 bytes
> >> > LZ4: compressed & decompressed 2535303543 bytes in 1.56s (1550.07MiB/s), mean compresion 99.60%, skipped 990 bytes
> >>
> >> Like your earlier comparison, this compares the wrong thing. If
> >> compression speed matters more than best compression ratio, you
> >> shouldn't use the default settings for xz. If you want to compare with
> >> LZ4, this benchmark should at least compare the equivalent of "xz -0".
> 
> > The comparison should be done for whatever setting is used in journald
> > and/or coredump.
> 
> No, The current defaults are irrelevant.
>
> The comparison should be done with values that indicate that
> introducing a new compression algorithm is needed. If xz *can* be
> instructed to reach acceptable results, we would better stick with it.
With preset=default (6):
XZ: compressed & decompressed 220536103 bytes in 60.03s (3.50MiB/s), ...

With 0:
XZ: compressed & decompressed 2535300963 bytes in 33.01s (73.25MiB/s), ...

LZ4: compressed & decompressed 2535303543 bytes in 1.31s (1838.75MiB/s), ...

Zbyszek