Details
-
Improvement
-
Status: Closed
-
Major
-
Resolution: Fixed
-
0.22.0
-
None
-
Reviewed
Description
Right now each connection into the datanode xceiver only processes one operation.
In the case that an operation leaves the stream in a well-defined state (eg a client reads to the end of a block successfully) the same connection could be reused for a second operation. This should improve random read performance significantly.
Attachments
Attachments
Issue Links
- is blocked by
-
HDFS-1001 DataXceiver and BlockReader disagree on when to send/recv CHECKSUM_OK
- Resolved
- is related to
-
HDFS-3359 DFSClient.close should close cached sockets
- Closed
-
HDFS-1599 Umbrella Jira for Improving HBASE support in HDFS
- Open
-
HBASE-3529 Add search to HBase
- Closed
- relates to
-
HDFS-2071 Use of isConnected() in DataXceiver is invalid
- Resolved
-
HDFS-380 support for persistent connections to improve random read performance.
- Resolved
-
HDFS-1148 Convert FSDataset to ReadWriteLock
- Patch Available