[
https://issues.apache.org/jira/browse/HBASE-16630?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15765587#comment-15765587
]
deepankar commented on HBASE-16630:
-----------------------------------
Sorry for the late reply, I looked into Stack's suggestion and tried to use for
this problem, the suggestion was more about improving the LRU part of the
eviction and I was not sure how that help in this case where the size bucket is
entirely full. There is already LRU by the means of accessCount and even though
that might not be fully optimal when compared to Stack's suggestion, in the
context of this problem the result from both the strategies will be same.
I looked into Ted Yu's suggestion last N blocks but the meta data overhead to
track for this case was too high. We tried to pinpoint the issue of re-eviction
int he heuristic used, with some debug statements / heap dump on real data but
we were not successful. So we have productionz-ed the attached version of patch
on more HBase clusters and are seeing some occasional issues but nothing
serious.
> Fragmentation in long running Bucket Cache
> ------------------------------------------
>
> Key: HBASE-16630
> URL: https://issues.apache.org/jira/browse/HBASE-16630
> Project: HBase
> Issue Type: Bug
> Components: BucketCache
> Affects Versions: 2.0.0, 1.1.6, 1.3.1, 1.2.3
> Reporter: deepankar
> Assignee: deepankar
> Priority: Critical
> Attachments: 16630-v2-suggest.patch, 16630-v3-suggest.patch,
> HBASE-16630-v2.patch, HBASE-16630-v3.patch, HBASE-16630.patch
>
>
> As we are running bucket cache for a long time in our system, we are
> observing cases where some nodes after some time does not fully utilize the
> bucket cache, in some cases it is even worse in the sense they get stuck at a
> value < 0.25 % of the bucket cache (DEFAULT_MEMORY_FACTOR as all our tables
> are configured in-memory for simplicity sake).
> We took a heap dump and analyzed what is happening and saw that is classic
> case of fragmentation, current implementation of BucketCache (mainly
> BucketAllocator) relies on the logic that fullyFreeBuckets are available for
> switching/adjusting cache usage between different bucketSizes . But once a
> compaction / bulkload happens and the blocks are evicted from a bucket size ,
> these are usually evicted from random places of the buckets of a bucketSize
> and thus locking the number of buckets associated with a bucketSize and in
> the worst case of the fragmentation we have seen some bucketSizes with
> occupancy ratio of < 10 % But they dont have any completelyFreeBuckets to
> share with the other bucketSize.
> Currently the existing eviction logic helps in the cases where cache used is
> more the MEMORY_FACTOR or MULTI_FACTOR and once those evictions are also
> done, the eviction (freeSpace function) will not evict anything and the cache
> utilization will be stuck at that value without any allocations for other
> required sizes.
> The fix for this we came up with is simple that we do deFragmentation (
> compaction) of the bucketSize and thus increasing the occupancy ratio and
> also freeing up the buckets to be fullyFree, this logic itself is not
> complicated as the bucketAllocator takes care of packing the blocks in the
> buckets, we need evict and re-allocate the blocks for all the BucketSizes
> that dont fit the criteria.
> I am attaching an initial patch just to give an idea of what we are thinking
> and I'll improve it based on the comments from the community.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
