On Friday Jan 17 2014, [email protected] wrote: > On Friday, January 17, 2014 3:07:02 AM UTC-5, Alan Schmitt wrote: >> >> Dirk-Jan C. Binnema <[email protected] <javascript:>> writes: >> >> >> Perhaps I'm overthinking this whole thing, but I'd be curious to hear >> >> some best-practices for people with >100k emails that don't want to >> >> wait 30 seconds for 'mu index' each time they update. >> > >> > 30 secs? That seems extremely long; after the first indexing, it only >> updates >> > messages that have changed, ie. >> >> > /home/djcb/Sources/mu/mu/mu index 0.32s user 0.48s system 82% cpu 0.974 >> total >> > >> > This is on an i5-2500 CPU @ 3.30GHz with 8G memory, and Fedora 20 >> > (kernel 3.12). >> >> I'm also seeing 25s, albeit with 250k messages: >> > >> 26.71 real 3.60 user 22.13 sys >> >> >> This is an a MacBook Air on OS X 10.9.1 (with SSD hard drive, 4G memory, >> 1.8 GHz Inter Core i7). >> >> > > I'm on a MacBook Pro from 2009, OS X 10.9.1, 5400 RPM 1 TB HD, 8 GB memory, > 2.26 GHz Core 2 Duo. > > I think the problem is the 5400 RPM disk. > > ~ $ time mu index > indexing messages under /Users/mankoff/Maildir [/Users/mankoff/.mu/xapian] > | processing mail; processed: 93075; updated/new: 0, cleaned-up: 0 > cleaning up messages [/Users/mankoff/.mu/xapian] > | processing mail; processed: 103498; updated/new: 0, cleaned-up: 0 > elapsed: 7 second(s), ~ 14785 msg/s > / processing mail; processed: 103498; updated/new: 0, cleaned-up: 0 > elapsed: 19 second(s), ~ 5447 msg/s
Perhaps something MacOS (or non-Linux) specific? I don't think a speed in harddrive could explain the order-of-magnitude difference in speed. I guess we should do some profiling on MacOS... Cheers, Dirk. -- 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.
