anaconda performance thoughts [Re: ANNOUNCE: Severn Test 2 Anaconda Updates Image Available]

[Matt Bernstein]

At 03:05 -0400 Ingo Molnar wrote:

>The biggest rpm we have currently is 40 MB, and the average package size
>is 1.4 MB. More than half of the total rpm size is in rpms that are
>smaller than 10 MB. With a 128 MB recommended RAM size this should be
>problem-free from a caching point of view. Maybe a special rule could be
>added, to not pre-cache RPMs with a size larger than RAM_size/4.

Maybe "pre-cache until size > [some threshold]" ?

>also, if the caching is done by copying the next rpm to HD while the rpm
>-i of the previous one is running then even if trashing happens, it will
>happen on the HD level, not on the CDROM level - and it's the CDROM that 
>is the fundamental bottleneck of installs.

Yes.. copy RPMs to a tmpfs volume. This would possibly depend on the 
target swap having been activated to prevent things going horribly wrong.

>ext2/ext3 only speeds up the HD access (by a very small amount). Also,
>ext2/ext3 mostly differs in CPU overhead not IO overhead - and the
>install-to-hd process is mostly limited by IO latencies (disk seeks).

I was thinking that ext3 writes couldn't return until metadata had been 
committed to real media, and that by aggressively write-caching ext2 might 
not need to. So, this is a different kind of I/O latency (one which may 
well be triggered by %post scripts). But if it was tried with no obvious 
performance gain, that is more pertinent than my guesswork!

Matt