[
https://issues.apache.org/jira/browse/HDFS-347?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13582695#comment-13582695
]
Colin Patrick McCabe commented on HDFS-347:
-------------------------------------------
bq. Why some clients don't want to do short-circuit? Could you give an example?
When using short-circuit local reads, you don't get all of the metrics that you
get with regular reads.
bq. LengthInputStream is a FilterInputStream. It is easy to return the
underlying input strem from FilterInputStream.
Can you be more specific about how you would like to do this?
bq. How to configure the existing short-circuit read (HDFS-2246) after the
patch?
On the DataNode side, the configuration parameters for old-style short-circuit
local reads haven't changed. On the client side, using old-style short-circuit
local reads is not possible. The server-side code is there only to provide
backwards compatibility. In other words, it is there to provide
interoperability between older clients and newer servers. We don't have to
maintain it forever, but I think we at least want the backwards compatibility
code in 2.0.x.
bq. I mean we might not need to compare version since we currently only has
one. If we have a new version in the future, the server could then detect the
old clients and fail them. Sound good?
My fear is that if we don't think through the compatibility issues now, we'll
have more bugs like HDFS-4506.
> DFS read performance suboptimal when client co-located on nodes with data
> -------------------------------------------------------------------------
>
> Key: HDFS-347
> URL: https://issues.apache.org/jira/browse/HDFS-347
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode, hdfs-client, performance
> Reporter: George Porter
> Assignee: Colin Patrick McCabe
> Attachments: 2013.01.28.design.pdf, 2013.01.31.consolidated2.patch,
> 2013.01.31.consolidated.patch, 2013.02.15.consolidated4.patch, all.tsv,
> BlockReaderLocal1.txt, full.patch, HADOOP-4801.1.patch, HADOOP-4801.2.patch,
> HADOOP-4801.3.patch, HDFS-347-016_cleaned.patch, HDFS-347.016.patch,
> HDFS-347.017.clean.patch, HDFS-347.017.patch, HDFS-347.018.clean.patch,
> HDFS-347.018.patch2, HDFS-347.019.patch, HDFS-347.020.patch,
> HDFS-347.021.patch, HDFS-347.022.patch, HDFS-347.024.patch,
> HDFS-347.025.patch, HDFS-347.026.patch, HDFS-347.027.patch,
> HDFS-347.029.patch, HDFS-347.030.patch, HDFS-347.033.patch,
> HDFS-347.035.patch, HDFS-347-branch-20-append.txt, hdfs-347-merge.txt,
> hdfs-347-merge.txt, hdfs-347-merge.txt, hdfs-347.png, hdfs-347.txt,
> local-reads-doc
>
>
> One of the major strategies Hadoop uses to get scalable data processing is to
> move the code to the data. However, putting the DFS client on the same
> physical node as the data blocks it acts on doesn't improve read performance
> as much as expected.
> After looking at Hadoop and O/S traces (via HADOOP-4049), I think the problem
> is due to the HDFS streaming protocol causing many more read I/O operations
> (iops) than necessary. Consider the case of a DFSClient fetching a 64 MB
> disk block from the DataNode process (running in a separate JVM) running on
> the same machine. The DataNode will satisfy the single disk block request by
> sending data back to the HDFS client in 64-KB chunks. In BlockSender.java,
> this is done in the sendChunk() method, relying on Java's transferTo()
> method. Depending on the host O/S and JVM implementation, transferTo() is
> implemented as either a sendfilev() syscall or a pair of mmap() and write().
> In either case, each chunk is read from the disk by issuing a separate I/O
> operation for each chunk. The result is that the single request for a 64-MB
> block ends up hitting the disk as over a thousand smaller requests for 64-KB
> each.
> Since the DFSClient runs in a different JVM and process than the DataNode,
> shuttling data from the disk to the DFSClient also results in context
> switches each time network packets get sent (in this case, the 64-kb chunk
> turns into a large number of 1500 byte packet send operations). Thus we see
> a large number of context switches for each block send operation.
> I'd like to get some feedback on the best way to address this, but I think
> providing a mechanism for a DFSClient to directly open data blocks that
> happen to be on the same machine. It could do this by examining the set of
> LocatedBlocks returned by the NameNode, marking those that should be resident
> on the local host. Since the DataNode and DFSClient (probably) share the
> same hadoop configuration, the DFSClient should be able to find the files
> holding the block data, and it could directly open them and send data back to
> the client. This would avoid the context switches imposed by the network
> layer, and would allow for much larger read buffers than 64KB, which should
> reduce the number of iops imposed by each read block operation.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
