[
https://issues.apache.org/jira/browse/HDFS-347?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12764953#action_12764953
]
Raghu Angadi commented on HDFS-347:
-----------------------------------
great to see progress on this.
- bq. This is does not apply to BlockSender, though, since that already forms
packets of (I think) 10 checksum chunks at a time.
** It is an issue for BlockSender as well since the limitation is imposed by
FSInputChecker. readChunk() only gives access to 512 bytes of user buffer to
implementation. Filed HADOOP-3205 quite sometime back. It would let BlockReader
avoid a copy as well.
- isLoopbackAddress()?
** Is this a temporary hack? Client sees non-loopback address even if the
datanode is local.
** In the current implementation you could first connect then check
(socket.localAddr == socket.remoteAddr) to decide go with local read.
- listening on "dfsclinet_clientname"
** What happens with multiple client reads?
** we could use "dfsclinet_some_rand_str_with_blockid".
** Instead, making datanode listen might have more advantages (like random
read latency mentioned below)
- random reads
** since random read bottleneck is connection latency and disk seeks, why do
you think this improves random read performance? This implementation has all
the latency overhead as before (Ignoring latency of connecting to local unix
socket, which might be negligible compared to a tcp connection).
** If we make datanode listen on unix domain, we could have real latency
improvement.
** extra thread is more than offset by threads avoided to read local data.
** does a typical HBase installation do predominantly do local reads?
- I suspect reporting bytes read by clients might be important issue to fix
properly.
** it could be an RPC to dn at the end of a proper termination.
** or if datanode is listening on unix domain, it could be sent over a unix
socket.
> 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
> Reporter: George Porter
> Assignee: Todd Lipcon
> Attachments: HADOOP-4801.1.patch, HADOOP-4801.2.patch,
> HADOOP-4801.3.patch, 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.
-
You can reply to this email to add a comment to the issue online.
