[
https://issues.apache.org/jira/browse/HDFS-7308?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14207309#comment-14207309
]
stack commented on HDFS-7308:
-----------------------------
Yeah, over in testing, I see [~szetszwo]'s calculated 65565 which is > power of
two 65536 and so we end up allocating buffers of near twice the size, 131072,
the next order of two whenever we have HDFS-7276 enabled. Thanks.
> DFSClient write packet size may > 64kB
> --------------------------------------
>
> Key: HDFS-7308
> URL: https://issues.apache.org/jira/browse/HDFS-7308
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: hdfs-client
> Reporter: Tsz Wo Nicholas Sze
> Assignee: Tsz Wo Nicholas Sze
> Priority: Minor
>
> In DFSOutputStream.computePacketChunkSize(..),
> {code}
> private void computePacketChunkSize(int psize, int csize) {
> final int chunkSize = csize + getChecksumSize();
> chunksPerPacket = Math.max(psize/chunkSize, 1);
> packetSize = chunkSize*chunksPerPacket;
> if (DFSClient.LOG.isDebugEnabled()) {
> ...
> }
> }
> {code}
> We have the following
> || variables || usual values ||
> | psize | dfsClient.getConf().writePacketSize = 64kB |
> | csize | bytesPerChecksum = 512B |
> | getChecksumSize(), i.e. CRC size | 32B |
> | chunkSize = csize + getChecksumSize() | 544B (not a power of two) |
> | psize/chunkSize | 120.47 |
> | chunksPerPacket = max(psize/chunkSize, 1) | 120 |
> | packetSize = chunkSize*chunksPerPacket (not including header) | 65280B |
> | PacketHeader.PKT_MAX_HEADER_LEN | 33B |
> | actual packet size | 65280 + 33 = *65313* < 65536 = 64k |
> It is fortunate that the usual packet size = 65313 < 64k although the
> calculation above does not guarantee it always happens (e.g. if
> PKT_MAX_HEADER_LEN=257, then actual packet size=65537 > 64k.) We should fix
> the computation in order to guarantee actual packet size < 64k.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
