hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Densel Santhmayor (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-18161) MultiHFileOutputFormat - comprehensive incremental load support
Date Tue, 06 Jun 2017 05:21:18 GMT

    [ https://issues.apache.org/jira/browse/HBASE-18161?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16038199#comment-16038199

Densel Santhmayor commented on HBASE-18161:

Please correct me if I'm mistaken here. 

>From initial analysis, HBASE-3727 did not have the ability to perform incremental load
and used the tablename as the mapper output key which severely limited the throughput for
tables with a large number of regions since there seems to be only a single reducer per table.

It seems that HBASE-16261 may solve this issue by writing its own partitioner and while the
tablename still needs to be sent as the mapper output key, a reducer is created for each splitkey
partition for all tables. However, a few things are still not clear to me: 
1. The configuration object seems to store only a *single set* of encodings of compression,
block encoding etc per family and this doesn't seem to be done per table. 
2. While HFileOutputFormat2 supports (for now) the KeyValueSortReducer, PutSortReducer, and
TextSortReducer, the new MultiTableHFileOutputFormat class only seems to support KeyValueSortReducer

My solution relies on upgrading HFileOutputFormat2 to support writing to multiple tables without
changing the underlying behaviour such as locality sensitivity and partitioning. I updated
all encoding functions to use as the key a combination of tablename and family, and the behaviour
of the new class - MultiHFileOutputFormat - as a public interface as well as providing helper
functions so that Mapper output keys (rowkey) can easily be transformed into a combination
of tablename and rowkey. With only these essential changes, any further changes to HFileOutputFormat2
will benefit users of MultiHFileOutputFormat with minor, if any, changes. For example, if
we need to support namespaces before the tablename, it should take only minor changes to a
few overridden functions - perhaps getOutputPath and getTableFamilyCombo 

In short, I aim to achieve a comprehensive solution to writing HFiles of multiple tables in
a single MR job, while preserving existing logic in HFileOutputFormat2 with minimal code changes,
effective client helper functions and meticulous error checking (If 3 tables are configured
initially through configureIncrementalLoad, then mapper output MUST only include either of
the 3 tables or an error is thrown when writing)

> MultiHFileOutputFormat - comprehensive incremental load support
> ---------------------------------------------------------------
>                 Key: HBASE-18161
>                 URL: https://issues.apache.org/jira/browse/HBASE-18161
>             Project: HBase
>          Issue Type: New Feature
>            Reporter: Densel Santhmayor
>            Priority: Minor
> h2. Introduction
> MapReduce currently supports the ability to write HBase records in bulk to HFiles for
a single table. The file(s) can then be uploaded to the relevant RegionServers information
with reasonable latency. This feature is useful to make a large set of data available for
queries at the same time as well as provides a way to efficiently process very large input
into HBase without affecting query latencies.
> There is, however, no support to write variations of the same record key to HFiles belonging
to multiple HBase tables from within the same MapReduce job.  
> h2. Goal
> The goal of this JIRA is to extend HFileOutputFormat2 to support writing to HFiles for
different tables within the same MapReduce job while single-table HFile features backwards-compatible.

> For our use case, we needed to write a record key to a smaller HBase table for quicker
access, and the same record key with a date appended to a larger table for longer term storage
with chronological access. Each of these tables would have different TTL and other settings
to support their respective access patterns. We also needed to be able to bulk write records
to multiple tables with different subsets of very large input as efficiently as possible.
Rather than run the MapReduce job multiple times (one for each table or record structure),
it would be useful to be able to parse the input a single time and write to multiple tables
> Additionally, we'd like to maintain backwards compatibility with the existing heavily-used
HFileOutputFormat2 interface to allow benefits such as locality sensitivity (that was introduced
long after we implemented support for multiple tables) to support both single table and multi
table hfile writes. 
> h2. Proposal
> * Backwards compatibility for existing single table support in HFileOutputFormat2 will
be maintained and in this case, mappers will need to emit the table rowkey as before. However,
a new class - MultiHFileOutputFormat - will provide a helper function to generate a rowkey
for mappers that prefixes the desired tablename to the existing rowkey as well as provides
configureIncrementalLoad support for multiple tables.
> * HFileOutputFormat2 will be updated in the following way:
> ** configureIncrementalLoad will now accept multiple table descriptor and region locator
pairs, analogous to the single pair currently accepted by HFileOutputFormat2. 
> ** Compression, Block Size, Bloom Type and Datablock settings PER column family that
are set in the Configuration object are now indexed and retrieved by tablename AND column
> ** getRegionStartKeys will now support multiple regionlocators and calculate split points
and therefore partitions collectively for all tables. Similarly, now the eventual number of
Reducers will be equal to the total number of partitions across all tables. 
> ** The RecordWriter class will be able to process rowkeys either with or without the
tablename prepended depending on how configureIncrementalLoad was configured with MultiHFileOutputFormat
or HFileOutputFormat2.
> * The use of MultiHFileOutputFormat will write the output into HFiles which will match
the output format of HFileOutputFormat2. However, while the default use case will keep the
existing directory structure with column family name as the directory and HFiles within that
directory, in the case of MultiHFileOutputFormat, it will output HFiles in the output directory
with the following relative paths: 
> {noformat}
>      --table1 
>        --family1 
>          --HFiles 
>      --table2 
>        --family1 
>        --family2 
>          --HFiles
> {noformat}
> This aims to be a comprehensive solution to the original tickets - HBASE-3727 and HBASE-16261.
Thanks to [~clayb] for his support.
> The patch will be attached shortly.

This message was sent by Atlassian JIRA

View raw message