[
https://issues.apache.org/jira/browse/HBASE-13408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14650518#comment-14650518
]
Eshcar Hillel commented on HBASE-13408:
---------------------------------------
Then how about we’ll make use of the FlushPolicy abstraction that is written so
nicely and is easy to extend ;-).
We can add to it a method selectStoresToCompact(), so that a flush process
manages 2 sets to reduce memory usage (1) stores to flush (2) stores to be
compacted. A store is in either of the two sets or in none, but not in both of
them. The decision whether reducing the memory usage is done by a flush or a
compaction depends on the store type and state.
In addition, we’ll add a method to the MemStore interface
doInmemoryCompaction(). In compacted memstore the implementation of this method
would be to push the active set into the compaction pipeline and invoke a
compaction.
With this solution the semantics of prepare-to-flush remains the same.
> HBase In-Memory Memstore Compaction
> -----------------------------------
>
> Key: HBASE-13408
> URL: https://issues.apache.org/jira/browse/HBASE-13408
> Project: HBase
> Issue Type: New Feature
> Reporter: Eshcar Hillel
> Attachments:
> HBaseIn-MemoryMemstoreCompactionDesignDocument-ver02.pdf,
> HBaseIn-MemoryMemstoreCompactionDesignDocument.pdf,
> InMemoryMemstoreCompactionEvaluationResults.pdf
>
>
> A store unit holds a column family in a region, where the memstore is its
> in-memory component. The memstore absorbs all updates to the store; from time
> to time these updates are flushed to a file on disk, where they are
> compacted. Unlike disk components, the memstore is not compacted until it is
> written to the filesystem and optionally to block-cache. This may result in
> underutilization of the memory due to duplicate entries per row, for example,
> when hot data is continuously updated.
> Generally, the faster the data is accumulated in memory, more flushes are
> triggered, the data sinks to disk more frequently, slowing down retrieval of
> data, even if very recent.
> In high-churn workloads, compacting the memstore can help maintain the data
> in memory, and thereby speed up data retrieval.
> We suggest a new compacted memstore with the following principles:
> 1. The data is kept in memory for as long as possible
> 2. Memstore data is either compacted or in process of being compacted
> 3. Allow a panic mode, which may interrupt an in-progress compaction and
> force a flush of part of the memstore.
> We suggest applying this optimization only to in-memory column families.
> A design document is attached.
> This feature was previously discussed in HBASE-5311.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
