[
https://issues.apache.org/jira/browse/HBASE-17339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15761917#comment-15761917
]
Ted Yu commented on HBASE-17339:
--------------------------------
bq. running 95% puts and 5% gets
The percentage of gets was not high.
Have you tried with 20% or more gets ?
bq. only 31%-35% of the operations require full scan
Was the above percentage w.r.t. the 5% gets ?
How much was the cache hit ratio ?
bq. reduction in cache accesses (70% less accesses)
Looks like the reduction means that the gets were for recently written data.
For cold data, how was latency affected ?
Thanks
> Scan-Memory-First Optimization
> ------------------------------
>
> Key: HBASE-17339
> URL: https://issues.apache.org/jira/browse/HBASE-17339
> Project: HBase
> Issue Type: Improvement
> Reporter: Eshcar Hillel
> Attachments: HBASE-17339-V01.patch
>
>
> The current implementation of a get operation (to retrieve values for a
> specific key) scans through all relevant stores of the region; for each store
> both memory components (memstores segments) and disk components (hfiles) are
> scanned in parallel.
> We suggest to apply an optimization that speculatively scans memory-only
> components first and only if the result is incomplete scans both memory and
> disk.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
