Re: benchmark between aranym and crest

[Michael Schmitz <schmitz@mail.biophys.uni-duesseldorf.de>]

> I have made a very small benchmark of aranym and crest:
> Here the results:
>
>               wget     tar zxf   rm -r      MPQS      galois
> =============================================================
> host-i386     3.313s    0.371s  0.036s     249 ms.     93 ms.
> host-i386     6.022s    0.209s  0.020s     248 ms.     93 ms.
> host-i386     2.795s    0.211s  0.021s     252 ms.     91 ms.
> host-amd64    3.552s    0.354s  0.047s     267 ms.     87 ms.
> host-amd64    2.756s    0.204s  0.033s     262 ms.     84 ms.
> host-amd64    2.819s    0.216s  0.020s     264 ms.     82 ms.
> crest         5.325s   26.827s  3.693s  28,160 ms. 10,850 ms.
> crest         5.152s   26.538s  4.015s  29,070 ms. 10,940 ms.
> crest         5.215s   24.745s  3.687s  28,590 ms. 10,750 ms.
> aranym-i386  26.820s   98.640s  8.550s  68,700 ms. 32,290 ms.
> aranym-i386  14.290s   96.580s  9.040s  68,280 ms. 32,770 ms.
> aranym-i386  10.590s   97.200s  9.660s  69,250 ms. 32,410 ms.
> aranym-amd64 13.940s  110.160s 10.230s  66,910 ms. 31,420 ms.
> aranym-amd64 10.240s  106.880s 10.010s  67,260 ms. 31,080 ms.
> aranym-amd64 11.820s  110.810s  9.880s  68,390 ms. 31,580 ms.
>
> So we see that aranym-amd64 and aranym-i386 are about the same
> speed, that crest is between 3 and 4 times faster and that the
> host is between 250 and 400 times faster.

Thanks for doing such an extensive benchmark. For testing package builds,
we really need Petr's disk access speedups ...

	Michael