From: Jaegeuk Kim Subject: Re: [PATCH] f2fs: optimize gc for better performance Date: Thu, 05 Sep 2013 08:50:09 +0900 Message-ID: <1378338609.2354.86.camel@kjgkr> References: <1377737323-11803-1-git-send-email-jinuxstyle@gmail.com> <1377777368.2354.46.camel@kjgkr> <52201A4F.9020808@gmail.com> <1378169118.2354.56.camel@kjgkr> <522677ED.8090203@gmail.com> <1378287635.2354.84.camel@kjgkr> <522732EB.4020103@gmail.com> Reply-To: jaegeuk.kim@samsung.com Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-reply-to: <522732EB.4020103@gmail.com> Sender: linux-kernel-owner@vger.kernel.org To: Jin Xu Cc: linux-f2fs-devel@lists.sourceforge.net, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: linux-f2fs-devel.lists.sourceforge.net Hi Jin, 2013-09-04 (=EC=88=98), 21:17 +0800, Jin Xu: > Hi Jaegeuk, >=20 > On 04/09/2013 17:40, Jaegeuk Kim wrote: > > Hi Jin, > > > > 2013-09-04 (=EC=88=98), 07:59 +0800, Jin Xu: > >> Hi Jaegeuk, > >> > >> On 03/09/2013 08:45, Jaegeuk Kim wrote: > >>> Hi Jin, > >>> > >>>> [...] > >>>>> > >>>>> It seems that we can obtain the performance gain just by settin= g the > >>>>> MAX_VICTIM_SEARCH to 4096, for example. > >>>>> So, how about just adding an ending criteria like below? > >>>>> > >>>> > >>>> I agree that we could get the performance improvement by simply > >>>> enlarging the MAX_VICTIM_SEARCH to 4096, but I am concerning the > >>>> scalability a little bit. Because it might always searching the = whole > >>>> bitmap in some cases, for example, when dirty segments is 4000 a= nd > >>>> total segments is 409600. > >>>>> [snip] > >>>> [...] > >>>>> > >>>>> if (p->max_search > MAX_VICTIM_SEARCH) > >>>>> p->max_search =3D MAX_VICTIM_SEARCH; > >>>>> > >>>> > >>>> The optimization does not apply to SSR mode. There has a reason. > >>>> As noticed in the test, when SSR selected the segments that have= most > >>>> garbage blocks, then when gc is needed, all the dirty segments m= ight > >>>> have very less garbage blocks, thus the gc overhead is high. Thi= s might > >>>> lead to performance degradation. So the patch does not change th= e > >>>> victim selection policy for SSR. > >>> > >>> I think it doesn't care. > >>> GC is only triggered during the direct node block allocation. > >>> What it means that we need to consider the number of GC triggers = where > >>> the GC triggers more frequently during the normal data allocation= than > >>> the node block allocation. > >>> So, I think it would not degrade performance significatly. > >>> > >>> BTW, could you show some numbers for this? > >>> Or could you test what I suggested? > >>> > >>> Thanks, > >>> > >> > >> I re-ran the test and got the following result: > >> > >> --------------------------------------- > >> 2GB SDHC > >> create 52023 files of size 32768 bytes > >> random re-write 100000 records of 4KB > >> --------------------------------------- > >> > >> | file creation (s) | rewrite time (s) | gc count | gc garbage blo= cks | > >> > >> no patch 341 4227 1174 174840 > >> patched 296 2995 634 109314 > >> patched (KIM) 324 2958 645 106682 > >> > >> In this test, it does not show the minor performance degradation c= aused > >> by applying the patch to SSR mode. Instead, the performance is a l= ittle > >> better with what you suggested. > >> > >> I agree that the performance degradation would not be significant = even > >> it does degrade. I ever saw the minor degradation in some workload= s, but > >> I didn't save the data. > >> > >> So, I agree that we can apply the patch to SSR mode as well. > >> > >> And do you still have concerns about the formula for calculating t= he # > >> of search? > > > > Thank you for the test. :) > > What I've concerned is that, if it is really important to get a vic= tim > > more accurately for the performance as you described, it doesn't ne= ed to > > calculate the number of searches IMO. Just let's select nr_dirty. W= hy > > not? > > Only the thing that we should consider is to handle the case where = the > > nr_dirty is too large. > > For this, we can just limit the # of searches to avoid performance > > degradation. > > > > Still actually, I'm not convincing the effectiveness of your formul= a. > > If possible, could you show it with numbers? >=20 > It's not easy to prove the effectiveness of the formula. It's just fo= r > eliminating my concern on the scalability of searching. Since it does > not matter much for the performance improvement, we can put it aside > and choose the simpler method as you suggested. >=20 > So, should I revise the patch based on what you suggested or will > you take care of it? Could you make a patch with your performance description and sumbit it again? Thanks a lot, --=20 Jaegeuk Kim Samsung