I just wanted to report that I got some new benchmark results for the
patched version of apt-file. This time the program was running in a
different system. Based on the results it turns out that the speed
improvements are not always warrantied, in fact in some systems the
execution times could be worse!
In systems with a low buffer cache (low RAM) using the original
compressed files yields faster results as the program has to read the
whole file each time from the disk. Although, if a system has enough
memory to cache the input files then processing the uncompressed files
is faster than the time need for reading the compressed input and
deflating it on the fly.
The benchmarks where performed in two different systems. The first one
has 2 Gigs of RAM and the patch is clearly showing a speed improvement.
The second system has 64 Megs and shows a speed degradation. It's hard
to quantify how much RAM is enough, although RAM is getting cheaper each
day and it's not uncommon to see laptops and desktops with Gigs of RAM.
Hopefully, this will not impact the acceptance of the patch.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
