[
https://issues.apache.org/jira/browse/HBASE-13260?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14518846#comment-14518846
]
Matteo Bertozzi commented on HBASE-13260:
-----------------------------------------
[~ndimiduk] for 1.1 is probably ok having a slow proc-store (keep in mind that
currently proc-wal is using hsync and the performance are really different from
the bench above done with hflush only, instead of 15sec for 1M ops I get 200sec
for 10k ops)
what I'd like to know from you is what do you have in mind for upgrading?
let say we can't get perf improvement on the region, and from the above "With
increased concurrency, the region performs worse (>10 threads)". for the
assignment we need a better throughput and shortcut on the replay will help too.
in my mind, I always had the different "stores" for groups of operations (e.g.
the DDL goes on its own wal, the assignment for sys-ns goes on its own wal and
so on). but using this we can even do DDL still uses the region-store and
assignment uses the proc-wal, and avoid any migration.
but I'd like to know what is your upgrade story in case the region patch can't
be improved. (can we get the full bench with 10, 30, 50 threads too, just for
reference?)
> Bootstrap Tables for fun and profit
> ------------------------------------
>
> Key: HBASE-13260
> URL: https://issues.apache.org/jira/browse/HBASE-13260
> Project: HBase
> Issue Type: Bug
> Reporter: Enis Soztutar
> Assignee: Enis Soztutar
> Fix For: 2.0.0, 1.1.0
>
> Attachments: hbase-13260_bench.patch, hbase-13260_prototype.patch
>
>
> Over at the ProcV2 discussions(HBASE-12439) and elsewhere I was mentioning an
> idea where we may want to use regular old regions to store/persist some data
> needed for HBase master to operate.
> We regularly use system tables for storing system data. acl, meta, namespace,
> quota are some examples. We also store the table state in meta now. Some data
> is persisted in zk only (replication peers and replication state, etc). We
> are moving away from zk as a permanent storage. As any self-respecting
> database does, we should store almost all of our data in HBase itself.
> However, we have an "availability" dependency between different kinds of
> data. For example all system tables need meta to be assigned first. All
> master operations need ns table to be assigned, etc.
> For at least two types of data, (1) procedure v2 states, (2) RS groups in
> HBASE-6721 we cannot depend on meta being assigned since "assignment" itself
> will depend on accessing this data. The solution in (1) is to implement a
> custom WAL format, and custom recover lease and WAL recovery. The solution in
> (2) is to have the table to store this data, but also cache it in zk for
> bootrapping initial assignments.
> For solving both of the above (and possible future use cases if any), I
> propose we add a "boostrap table" concept, which is:
> - A set of predefined tables hosted in a separate dir in HDFS.
> - A table is only 1 region, not splittable
> - Not assigned through regular assignment
> - Hosted only on 1 server (typically master)
> - Has a dedicated WAL.
> - A service does WAL recovery + fencing for these tables.
> This has the benefit of using a region to keep the data, but frees us to
> re-implement caching and we can use the same WAL / Memstore / Recovery
> mechanisms that are battle-tested.
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
