hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anoop Sam John (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17072) CPU usage starts to climb up to 90-100% when using G1GC; purge ThreadLocal usage
Date Thu, 24 Nov 2016 07:49:59 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17072?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15692551#comment-15692551
] 

Anoop Sam John commented on HBASE-17072:
----------------------------------------


Oh you mean we will be always passing the block on disk size so no way we need the prefetched
header. My doubt was that only.. Reading code more, I can see HFileScanner#seekTo() only passing
the block on disk size as -1. Ya that is the 1st block.
Other places we seems to pass some size.(block.getNextBlockOnDiskSize())
{code}
/**
   * The on-disk size of the next block, including the header and checksums if present, obtained
by
   * peeking into the first {@link HConstants#HFILEBLOCK_HEADER_SIZE} bytes of the next block's
   * header, or UNSET if unknown.
   *
   * Blocks try to carry the size of the next block to read in this data member. They will
even have
   * this value when served from cache. Could save a seek in the case where we are iterating
through
   * a file and some of the blocks come from cache. If from cache, then having this info to
hand
   * will save us doing a seek to read the header so we can read the body of a block.
   * TODO: see how effective this is at saving seeks.
   */
  private int nextBlockOnDiskSize = UNSET;
  {code}
  So it says the state we get by seeing next block's header bytes. So we can not really avoid
it?  Or this comment is wrong now?

> CPU usage starts to climb up to 90-100% when using G1GC; purge ThreadLocal usage
> --------------------------------------------------------------------------------
>
>                 Key: HBASE-17072
>                 URL: https://issues.apache.org/jira/browse/HBASE-17072
>             Project: HBase
>          Issue Type: Bug
>          Components: Performance, regionserver
>    Affects Versions: 1.0.0, 2.0.0, 1.2.0
>            Reporter: Eiichi Sato
>         Attachments: HBASE-17072.master.001.patch, HBASE-17072.master.002.patch, HBASE-17072.master.003.patch,
HBASE-17072.master.004.patch, HBASE-17072.master.005.patch, disable-block-header-cache.patch,
mat-threadlocals.png, mat-threads.png, metrics.png, slave1.svg, slave2.svg, slave3.svg, slave4.svg
>
>
> h5. Problem
> CPU usage of a region server in our CDH 5.4.5 cluster, at some point, starts to gradually
get higher up to nearly 90-100% when using G1GC.  We've also run into this problem on CDH
5.7.3 and CDH 5.8.2.
> In our production cluster, it normally takes a few weeks for this to happen after restarting
a RS.  We reproduced this on our test cluster and attached the results.  Please note that,
to make it easy to reproduce, we did some "anti-tuning" on a table when running tests.
> In metrics.png, soon after we started running some workloads against a test cluster (CDH
5.8.2) at about 7 p.m. CPU usage of the two RSs started to rise.  Flame Graphs (slave1.svg
to slave4.svg) are generated from jstack dumps of each RS process around 10:30 a.m. the next
day.
> After investigating heapdumps from another occurrence on a test cluster running CDH 5.7.3,
we found that the ThreadLocalMap contain a lot of contiguous entries of {{HFileBlock$PrefetchedHeader}}
probably due to primary clustering.  This caused more loops in {{ThreadLocalMap#expungeStaleEntries()}},
consuming a certain amount of CPU time.  What is worse is that the method is called from RPC
metrics code, which means even a small amount of per-RPC time soon adds up to a huge amount
of CPU time.
> This is very similar to the issue in HBASE-16616, but we have many {{HFileBlock$PrefetchedHeader}}
not only {{Counter$IndexHolder}} instances.  Here are some OQL counts from Eclipse Memory
Analyzer (MAT).  This shows a number of ThreadLocal instances in the ThreadLocalMap of a single
handler thread.
> {code}
> SELECT *
> FROM OBJECTS (SELECT AS RETAINED SET OBJECTS value
> 			  FROM OBJECTS 0x4ee380430) obj
> WHERE obj.@clazz.@name = "org.apache.hadoop.hbase.io.hfile.HFileBlock$PrefetchedHeader"
> #=> 10980 instances
> {code}
> {code}
> SELECT *
> FROM OBJECTS (SELECT AS RETAINED SET OBJECTS value
> 			  FROM OBJECTS 0x4ee380430) obj
> WHERE obj.@clazz.@name = "org.apache.hadoop.hbase.util.Counter$IndexHolder"
> #=> 2052 instances
> {code}
> Although as described in HBASE-16616 this somewhat seems to be an issue in G1GC side
regarding weakly-reachable objects, we should keep ThreadLocal usage minimal and avoid creating
an indefinite number (in this case, a number of HFiles) of ThreadLocal instances.
> HBASE-16146 removes ThreadLocals from the RPC metrics code.  That may solve the issue
(I just saw the patch, never tested it at all), but the {{HFileBlock$PrefetchedHeader}} are
still there in the ThreadLocalMap, which may cause issues in the future again.
> h5. Our Solution
> We simply removed the whole {{HFileBlock$PrefetchedHeader}} caching and fortunately we
didn't notice any performance degradation for our production workloads.
> Because the PrefetchedHeader caching uses ThreadLocal and because RPCs are handled randomly
in any of the handlers, small Get or small Scan RPCs do not benefit from the caching (See
HBASE-10676 and HBASE-11402 for the details).  Probably, we need to see how well reads are
saved by the caching for large Scan or Get RPCs and especially for compactions if we really
remove the caching. It's probably better if we can remove ThreadLocals without breaking the
current caching behavior.
> FWIW, I'm attaching the patch we applied. It's for CDH 5.4.5.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message