[
https://issues.apache.org/jira/browse/HBASE-11165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13999605#comment-13999605
]
Mikhail Antonov commented on HBASE-11165:
-----------------------------------------
Before we had META hosted on master, I'd have said that this situation is
different from NN - as for NN it's kind of known that 1Gb or RAM can hold ~1M
inodes or so and whole FSImage must be in memory. With META tied to Master it's
kind of same issue? Does someone have estimates - how many memory on avg. are
needed for META table for each, say, 10k or 100k regions?
Do we have some list of bottlenecks on this way? I remember from the hackathon
@ Salesforce that one of important bottlenecks is number of splits/assignment
ops / sec (do I recall correctly [~toffer]?)
> Scaling so cluster can host 1M regions and beyond (50M regions?)
> ----------------------------------------------------------------
>
> Key: HBASE-11165
> URL: https://issues.apache.org/jira/browse/HBASE-11165
> Project: HBase
> Issue Type: Brainstorming
> Reporter: stack
>
> This discussion issue comes out of "Co-locate Meta And Master HBASE-10569"
> and comments on the doc posted there.
> A user -- our Francis Liu -- needs to be able to scale a cluster to do 1M
> regions maybe even 50M later. This issue is about discussing how we will do
> that (or if not 50M on a cluster, how otherwise we can attain same end).
> More detail to follow.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
