[
https://issues.apache.org/jira/browse/HDFS-8722?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Kihwal Lee updated HDFS-8722:
-----------------------------
Attachment: HDFS-8722.patch
The patch distinguishes the conditions for recalculating CRC from overwriting
the last checksum. If the on-disk data is not chunk-aligned, the checksum is
overwritten, but checksum recalculation will be done only when the incoming
data does not start at the last chunk boundary. Otherwise, the client supplied
chunk checksum will be used for overwriting the last partial chunk checksum.
> Optimize datanode writes for small writes and flushes
> -----------------------------------------------------
>
> Key: HDFS-8722
> URL: https://issues.apache.org/jira/browse/HDFS-8722
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Kihwal Lee
> Assignee: Kihwal Lee
> Priority: Critical
> Attachments: HDFS-8722.patch
>
>
> After the data corruption fix by HDFS-4660, the CRC recalculation for partial
> chunk is executed more frequently, if the client repeats writing few bytes
> and calling hflush/hsync. This is because the generic logic forces CRC
> recalculation if on-disk data is not CRC chunk aligned. Prior to HDFS-4660,
> datanode blindly accepted whatever CRC client provided, if the incoming data
> is chunk-aligned. This was the source of the corruption.
> We can still optimize for the most common case where a client is repeatedly
> writing small number of bytes followed by hflush/hsync with no pipeline
> recovery or append, by allowing the previous behavior for this specific case.
> If the incoming data has a duplicate portion and that is at the last
> chunk-boundary before the partial chunk on disk, datanode can use the
> checksum supplied by the client without redoing the checksum on its own.
> This reduces disk reads as well as CPU load for the checksum calculation.
> If the incoming packet data goes back further than the last on-disk chunk
> boundary, datanode will still do a recalculation, but this occurs rarely
> during pipeline recoveries. Thus the optimization for this specific case
> should be sufficient to speed up the vast majority of cases.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
