On Fri, Aug 31, 2012 at 5:58 PM, Mark Phippard <markp...@gmail.com> wrote: > On Fri, Aug 31, 2012 at 8:03 AM, Ivan Zhakov <i...@visualsvn.com> wrote: >> >> On Thu, Aug 30, 2012 at 10:43 PM, C. Michael Pilato <cmpil...@collab.net> >> wrote: >> > On 08/30/2012 10:45 AM, Ivan Zhakov wrote: >> >> On Thu, Aug 30, 2012 at 4:56 PM, C. Michael Pilato >> >> <cmpil...@collab.net> wrote: >> >>> Theoretically, though, it seems reasonable that my approach would have >> >>> the >> >>> distinct non-feature of potentially having the client caching the >> >>> properties >> >>> for an entire tree in memory, just waiting for a place to put them. >> >> >> >> It should not happen in current ra_serf editor implementation: parsing >> >> REPORT will be paused when active_fetchs + active_propfinds greater >> >> than 1000. See REQUEST_COUNT_TO_PAUSE/REQUEST_COUNT_TO_RESUME >> >> constants in libsvn_ra_serf/update.c >> > >> > Excellent! Thanks for pointing that out, Ivan. >> > >> I've just tested your change and I want to say that it's really cool: >> switching between serf trunk and 0.3.x branch is just 5 (five!) >> requests and 27k of traffic. I think it's my most wanted change in svn >> 1.8.x :) > > > Do you have your own internal server built from trunk that you are using for > testing or do you also see benefits with current servers? Only asking > because you mentioned serf so I am wondering if you are talking about the > real serf repository or your own mirror. > I'm using my own serf repository mirror. Svn 1.8.x is also faster when working with svn 1.7.x servers (because it doesn't fetch files content) but with 1.8.x it's really fast.
-- Ivan Zhakov
