On Wednesday,2009-09-23, at 2:17 , Thomas Delaet wrote: >> IIRC the one which currently works the best is impl_c. > > That is the only one I got up and running with a few > modifications ... ;-)
Yay! That's good news! Would you please give us the patch of your modifications? > I use the Postmark benchmark, which needs a POSIX interface, so I > have no other option in that area. In my project, I also test the > Wuala system and their interface uses a local cache to improve > performance. Maybe this is something Tahoe can consider too? Yes. It is under consideration -- tickets #300, #316, #465, #606. I'm leery of adding caching to Tahoe-LAFS proper because caching is usually a tradeoff of gaining performance at a cost of correctness, and the apps layered atop Tahoe-LAFS can know better than Tahoe-LAFS can which of those tradeoffs are wins for them. I guess it is always safe to cache immutable files, but is that what people want? Hopefully your benchmarks will give us some clue. > Kind Regards and thanks a lot for your help Aren't you going to tell us what you learned? Regards, Zooko tickets mentioned in this letter: http://allmydata.org/trac/tahoe/ticket/300 # macfuse: need some sort of caching http://allmydata.org/trac/tahoe/ticket/316 # add caching to tahoe proper? http://allmydata.org/trac/tahoe/ticket/465 # add a mutable-file cache http://allmydata.org/trac/tahoe/ticket/606 # backupdb: add directory cache _______________________________________________ tahoe-dev mailing list [email protected] http://allmydata.org/cgi-bin/mailman/listinfo/tahoe-dev
