Hello! I think that good robust approach is to start background thread which will try to compact pages and remove unneeded ones. It should only be active when system is reasonably idle, or if there's severe fragmentation problem.
However, I am aware that implementing such heurestical cleaner is a challenging task. Regards, -- Ilya Kasnacheev пт, 4 окт. 2019 г. в 19:38, Alexey Goncharuk <alexey.goncha...@gmail.com>: > Maxim, > > Having a cluster-wide lock for a cache does not improve availability of the > solution. A user cannot defragment a cache if the cache is involved in a > mission-critical operation, so having a lock on such a cache is equivalent > to the whole cluster shutdown. > > We should decide between either a single offline node or a more complex > fully online solution. > > пт, 4 окт. 2019 г. в 11:55, Maxim Muzafarov <mmu...@apache.org>: > > > Igniters, > > > > This thread seems to be endless, but we if some kind of cache group > > distributed write lock (exclusive for some of the internal Ignite > > process) will be introduced? I think it will help to solve a batch of > > problems, like: > > > > 1. defragmentation of all cache group partitions on the local node > > without concurrent updates. > > 2. improve data loading with data streamer isolation mode [1]. It > > seems we should not allow concurrent updates to cache if we on `fast > > data load` step. > > 3. recovery from a snapshot without cache stop\start actions > > > > > > [1] https://issues.apache.org/jira/browse/IGNITE-11793 > > > > On Thu, 3 Oct 2019 at 22:50, Sergey Kozlov <skoz...@gridgain.com> wrote: > > > > > > Hi > > > > > > I'm not sure that node offline is a best way to do that. > > > Cons: > > > - different caches may have different defragmentation but we force to > > stop > > > whole node > > > - offline node is a maintenance operation will require to add +1 > backup > > to > > > reduce the risk of data loss > > > - baseline auto adjustment? > > > - impact to index rebuild? > > > - cache configuration changes (or destroy) during node offline > > > > > > What about other ways without node stop? E.g. make cache group on a > node > > > offline? Add *defrag <cache_group> *command to control.sh to force > start > > > rebalance internally in the node with expected impact to performance. > > > > > > > > > > > > On Thu, Oct 3, 2019 at 12:08 PM Anton Vinogradov <a...@apache.org> > wrote: > > > > > > > Alexey, > > > > As for me, it does not matter will it be IEP, umbrella or a single > > issue. > > > > The most important thing is Assignee :) > > > > > > > > On Thu, Oct 3, 2019 at 11:59 AM Alexey Goncharuk < > > > > alexey.goncha...@gmail.com> > > > > wrote: > > > > > > > > > Anton, do you think we should file a single ticket for this or > > should we > > > > go > > > > > with an IEP? As of now, the change does not look big enough for an > > IEP > > > > for > > > > > me. > > > > > > > > > > чт, 3 окт. 2019 г. в 11:18, Anton Vinogradov <a...@apache.org>: > > > > > > > > > > > Alexey, > > > > > > > > > > > > Sounds good to me. > > > > > > > > > > > > On Thu, Oct 3, 2019 at 10:51 AM Alexey Goncharuk < > > > > > > alexey.goncha...@gmail.com> > > > > > > wrote: > > > > > > > > > > > > > Anton, > > > > > > > > > > > > > > Switching a partition to and from the SHRINKING state will > > require > > > > > > > intricate synchronizations in order to properly determine the > > start > > > > > > > position for historical rebalance without PME. > > > > > > > > > > > > > > I would still go with an offline-node approach, but instead of > > > > cleaning > > > > > > the > > > > > > > persistence, we can do effective defragmentation when the node > is > > > > > offline > > > > > > > because we are sure that there is no concurrent load. After the > > > > > > > defragmentation completes, we bring the node back to the > cluster > > and > > > > > > > historical rebalance will kick in automatically. It will still > > > > require > > > > > > > manual node restarts, but since the data is not removed, there > > are no > > > > > > > additional risks. Also, this will be an excellent solution for > > those > > > > > who > > > > > > > can afford downtime and execute the defragment command on all > > nodes > > > > in > > > > > > the > > > > > > > cluster simultaneously - this will be the fastest way possible. > > > > > > > > > > > > > > --AG > > > > > > > > > > > > > > пн, 30 сент. 2019 г. в 09:29, Anton Vinogradov <a...@apache.org > >: > > > > > > > > > > > > > > > Alexei, > > > > > > > > >> stopping fragmented node and removing partition data, then > > > > > starting > > > > > > it > > > > > > > > again > > > > > > > > > > > > > > > > That's exactly what we're doing to solve the fragmentation > > issue. > > > > > > > > The problem here is that we have to perform N/B > > restart-rebalance > > > > > > > > operations (N - cluster size, B - backups count) and it takes > > a lot > > > > > of > > > > > > > time > > > > > > > > with risks to lose the data. > > > > > > > > > > > > > > > > On Fri, Sep 27, 2019 at 5:49 PM Alexei Scherbakov < > > > > > > > > alexey.scherbak...@gmail.com> wrote: > > > > > > > > > > > > > > > > > Probably this should be allowed to do using public API, > > actually > > > > > this > > > > > > > is > > > > > > > > > same as manual rebalancing. > > > > > > > > > > > > > > > > > > пт, 27 сент. 2019 г. в 17:40, Alexei Scherbakov < > > > > > > > > > alexey.scherbak...@gmail.com>: > > > > > > > > > > > > > > > > > > > The poor man's solution for the problem would be stopping > > > > > > fragmented > > > > > > > > node > > > > > > > > > > and removing partition data, then starting it again > > allowing > > > > full > > > > > > > state > > > > > > > > > > transfer already without deletes. > > > > > > > > > > Rinse and repeat for all owners. > > > > > > > > > > > > > > > > > > > > Anton Vinogradov, would this work for you as workaround ? > > > > > > > > > > > > > > > > > > > > чт, 19 сент. 2019 г. в 13:03, Anton Vinogradov < > > a...@apache.org > > > > >: > > > > > > > > > > > > > > > > > > > >> Alexey, > > > > > > > > > >> > > > > > > > > > >> Let's combine your and Ivan's proposals. > > > > > > > > > >> > > > > > > > > > >> >> vacuum command, which acquires exclusive table lock, > > so no > > > > > > > > concurrent > > > > > > > > > >> activities on the table are possible. > > > > > > > > > >> and > > > > > > > > > >> >> Could the problem be solved by stopping a node which > > needs > > > > to > > > > > > be > > > > > > > > > >> defragmented, clearing persistence files and restarting > > the > > > > > node? > > > > > > > > > >> >> After rebalancing the node will receive all data back > > > > without > > > > > > > > > >> fragmentation. > > > > > > > > > >> > > > > > > > > > >> How about to have special partition state SHRINKING? > > > > > > > > > >> This state should mean that partition unavailable for > > reads > > > > and > > > > > > > > updates > > > > > > > > > >> but > > > > > > > > > >> should keep it's update-counters and should not be > marked > > as > > > > > lost, > > > > > > > > > renting > > > > > > > > > >> or evicted. > > > > > > > > > >> At this state we able to iterate over the partition and > > apply > > > > > it's > > > > > > > > > entries > > > > > > > > > >> to another file in a compact way. > > > > > > > > > >> Indices should be updated during the copy-on-shrink > > procedure > > > > or > > > > > > at > > > > > > > > the > > > > > > > > > >> shrink completion. > > > > > > > > > >> Once shrank file is ready we should replace the original > > > > > partition > > > > > > > > file > > > > > > > > > >> with it and mark it as MOVING which will start the > > historical > > > > > > > > rebalance. > > > > > > > > > >> Shrinking should be performed during the low activity > > periods, > > > > > but > > > > > > > > even > > > > > > > > > in > > > > > > > > > >> case we found that activity was high and historical > > rebalance > > > > is > > > > > > not > > > > > > > > > >> suitable we may just remove the file and use regular > > rebalance > > > > > to > > > > > > > > > restore > > > > > > > > > >> the partition (this will also lead to shrink). > > > > > > > > > >> > > > > > > > > > >> BTW, seems, we able to implement partition shrink in a > > cheap > > > > > way. > > > > > > > > > >> We may just use rebalancing code to apply fat > partition's > > > > > entries > > > > > > to > > > > > > > > the > > > > > > > > > >> new file. > > > > > > > > > >> So, 3 stages here: local rebalance, indices update and > > global > > > > > > > > historical > > > > > > > > > >> rebalance. > > > > > > > > > >> > > > > > > > > > >> On Thu, Sep 19, 2019 at 11:43 AM Alexey Goncharuk < > > > > > > > > > >> alexey.goncha...@gmail.com> wrote: > > > > > > > > > >> > > > > > > > > > >> > Anton, > > > > > > > > > >> > > > > > > > > > > >> > > > > > > > > > > >> > > >> The solution which Anton suggested does not look > > easy > > > > > > > because > > > > > > > > it > > > > > > > > > >> will > > > > > > > > > >> > > most likely significantly hurt performance > > > > > > > > > >> > > Mostly agree here, but what drop do we expect? What > > price > > > > do > > > > > > we > > > > > > > > > ready > > > > > > > > > >> to > > > > > > > > > >> > > pay? > > > > > > > > > >> > > Not sure, but seems some vendors ready to pay, for > > > > example, > > > > > 5% > > > > > > > > drop > > > > > > > > > >> for > > > > > > > > > >> > > this. > > > > > > > > > >> > > > > > > > > > > >> > 5% may be a big drop for some use-cases, so I think we > > > > should > > > > > > look > > > > > > > > at > > > > > > > > > >> how > > > > > > > > > >> > to improve performance, not how to make it worse. > > > > > > > > > >> > > > > > > > > > > >> > > > > > > > > > > >> > > > > > > > > > > > >> > > >> it is hard to maintain a data structure to choose > > "page > > > > > > from > > > > > > > > > >> free-list > > > > > > > > > >> > > with enough space closest to the beginning of the > > file". > > > > > > > > > >> > > We can just split each free-list bucket to the > couple > > and > > > > > use > > > > > > > > first > > > > > > > > > >> for > > > > > > > > > >> > > pages in the first half of the file and the second > > for the > > > > > > last. > > > > > > > > > >> > > Only two buckets required here since, during the > file > > > > > shrink, > > > > > > > > first > > > > > > > > > >> > > bucket's window will be shrank too. > > > > > > > > > >> > > Seems, this give us the same price on put, just use > > the > > > > > first > > > > > > > > bucket > > > > > > > > > >> in > > > > > > > > > >> > > case it's not empty. > > > > > > > > > >> > > Remove price (with merge) will be increased, of > > course. > > > > > > > > > >> > > > > > > > > > > > >> > > The compromise solution is to have priority put (to > > the > > > > > first > > > > > > > path > > > > > > > > > of > > > > > > > > > >> the > > > > > > > > > >> > > file), with keeping removal as is, and schedulable > > > > per-page > > > > > > > > > migration > > > > > > > > > >> for > > > > > > > > > >> > > the rest of the data during the low activity period. > > > > > > > > > >> > > > > > > > > > > > >> > Free lists are large and slow by themselves, it is > > expensive > > > > > to > > > > > > > > > >> checkpoint > > > > > > > > > >> > and read them on start, so as a long-term solution I > > would > > > > > look > > > > > > > into > > > > > > > > > >> > removing them. Moreover, not sure if adding yet > another > > > > > > background > > > > > > > > > >> process > > > > > > > > > >> > will improve the codebase reliability and simplicity. > > > > > > > > > >> > > > > > > > > > > >> > If we want to go the hard path, I would look at free > > page > > > > > > tracking > > > > > > > > > >> bitmap - > > > > > > > > > >> > a special bitmask page, where each page in an adjacent > > block > > > > > is > > > > > > > > marked > > > > > > > > > >> as 0 > > > > > > > > > >> > if it has free space more than a certain configurable > > > > > threshold > > > > > > > > (say, > > > > > > > > > >> 80%) > > > > > > > > > >> > - free, and 1 if less (full). Some vendors have > > successfully > > > > > > > > > implemented > > > > > > > > > >> > this approach, which looks much more promising, but > > harder > > > > to > > > > > > > > > implement. > > > > > > > > > >> > > > > > > > > > > >> > --AG > > > > > > > > > >> > > > > > > > > > > >> > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > > > > > > > > > > > > > > Best regards, > > > > > > > > > > Alexei Scherbakov > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > > > > > > > > > > > > Best regards, > > > > > > > > > Alexei Scherbakov > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > Sergey Kozlov > > > GridGain Systems > > > www.gridgain.com > > >
