On Tuesday Jan 21 2014, [email protected] wrote:

> I'd say that's the difference between an ssd and a 5400rpm platter disk.
>
> I have seen this deteriorate with lots of files, but then the difference is
> an order of magnitude.
>
> I could of course be mistaken and I'm happy to do more testing...
> Just don't know what really.
> On 21 Jan 2014 15:05, "Ken Mankoff" <[email protected]> wrote:
>
>>
>> Rules what out? It took 2x as long (10 seconds) just to list the
>> files. Seems like that might be part of the problem.
>>


>>>>> related.
>>>>> $ time ls -R1 ~/Maildir
>>>>>

I guess for the filesystem performance, we should discard the output;
and free various file-system caches; not sure how to do that on MacOS,
but on Linux ((for my 50K messages):

% sudo sh -c 'sync && echo 3 > /proc/sys/vm/drop_caches'
% time ls -R1 ~/Maildir > /dev/null
0.28s user 0.07s system 16% cpu 2.056 total

if we do it again (with a warm cache):
% time ls -R1 ~/Maildir > /dev/null
0.18s user 0.03s system 99% cpu 0.213 total


>>>>  MBP 2.26 GHz Core Due 2
>>>> 8 GB RAM
>>>> 5400 RPM drive
>>>>
>>>> | real | 0m10.571s |
>>>> | user | 0m1.134s  |
>>>> | sys  | 0m0.520s  |
>>>>
>>>> ~100k files.



-- 
Dirk-Jan C. Binnema                  Helsinki, Finland
e:[email protected]           w:www.djcbsoftware.nl
pgp: D09C E664 897D 7D39 5047 A178 E96A C7A1 017D DA3C

-- 
You received this message because you are subscribed to the Google Groups 
"mu-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to