HBASE-6701 Revisit thrust of paragraph on splitting (Misty Stanley-Jones)

This commit is contained in:
Michael Stack 2014-06-02 10:02:23 -07:00
parent db9cb9ca08
commit 6fdf737171
1 changed files with 8 additions and 8 deletions

View File

@ -1374,14 +1374,14 @@ index e70ebc6..96f8c27 100644
<para>See <xref linkend="master.processes.loadbalancer" /> for more information on the LoadBalancer. <para>See <xref linkend="master.processes.loadbalancer" /> for more information on the LoadBalancer.
</para> </para>
</section> </section>
<section xml:id="disabling.blockcache"><title>Disabling Blockcache</title> <section xml:id="disabling.blockcache"><title>Disabling Blockcache</title>
<para>Do not turn off block cache (You'd do it by setting <varname>hbase.block.cache.size</varname> to zero). <para>Do not turn off block cache (You'd do it by setting <varname>hbase.block.cache.size</varname> to zero).
Currently we do not do well if you do this because the regionserver will spend all its time loading hfile Currently we do not do well if you do this because the regionserver will spend all its time loading hfile
indices over and over again. In fact, in later versions of HBase, it is not possible to disable the indices over and over again. If your working set it such that block cache does you no good, at least
block cache completely. size the block cache such that hfile indices will stay up in the cache (you can get a rough idea
HBase will cache meta blocks -- the INDEX and BLOOM blocks -- even if the block cache on the size you need by surveying regionserver UIs; you'll see index block size accounted near the
is disabled.</para> top of the webpage).</para>
</section> </section>
<section xml:id="nagles"> <section xml:id="nagles">
<title><link xlink:href="http://en.wikipedia.org/wiki/Nagle's_algorithm">Nagle's</link> or the small package problem</title> <title><link xlink:href="http://en.wikipedia.org/wiki/Nagle's_algorithm">Nagle's</link> or the small package problem</title>
<para>If a big 40ms or so occasional delay is seen in operations against HBase, <para>If a big 40ms or so occasional delay is seen in operations against HBase,