Repository: accumulo
Updated Branches:
  refs/heads/asf-site b1bf6c22e -> 45d81dfc4
  refs/heads/gh-pages a9b0fda04 -> 6538290f3


Minor updates to 1.7.2 release notes.


Project: http://git-wip-us.apache.org/repos/asf/accumulo/repo
Commit: http://git-wip-us.apache.org/repos/asf/accumulo/commit/6538290f
Tree: http://git-wip-us.apache.org/repos/asf/accumulo/tree/6538290f
Diff: http://git-wip-us.apache.org/repos/asf/accumulo/diff/6538290f

Branch: refs/heads/gh-pages
Commit: 6538290f370ab48ad59526c87b5733c60a31cafa
Parents: a9b0fda
Author: Josh Elser <els...@apache.org>
Authored: Thu Jun 23 10:55:04 2016 -0400
Committer: Josh Elser <els...@apache.org>
Committed: Thu Jun 23 10:55:04 2016 -0400

----------------------------------------------------------------------
 release_notes/1.7.2.md | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/accumulo/blob/6538290f/release_notes/1.7.2.md
----------------------------------------------------------------------
diff --git a/release_notes/1.7.2.md b/release_notes/1.7.2.md
index 5ffce6f..eebdc70 100644
--- a/release_notes/1.7.2.md
+++ b/release_notes/1.7.2.md
@@ -31,6 +31,10 @@ There was a bug ([ACCUMULO-4148][ACCUMULO-4148]) where 
multiple put calls with i
 
 An improvement was introduced to allow a max age before WAL files would be 
automatically rolled. Without a max age, they could stay open for writing 
indefinitely, blocking the Hadoop DataNode decomissioning process. For more 
information, see [ACCUMULO-4004][ACCUMULO-4004].
 
+### Remove copy of cached RFile index blocks
+
+Accumulo maintains an cache for file blocks in-memory as a performance 
optimization. This can be done safely because Accumulo RFiles are immutable, 
thus their blocks are also immutable. There are two types of these blocks: 
index and data blocks. Index blocks refer to the b-tree style index inside of 
each Accumulo RFile, while data blocks contain the sorted Key-Value pairs. In 
previous versions, when Accumulo extracted an Index block from the in-memory 
cache, it would copy the data. [ACCUMULO-4164][ACCUMULO-4164] removes this 
unnecessary copy as the contents are immutable and can be passed by reference. 
Ensuring that the Index blocks are not copied when accessed from the cache is a 
big performance gain at the file-access level.
+
 ### Minor performance improvements.
 
 Tablet servers would previously always hsync at the start of a minor 
compaction, causing delays in the write pipeline. These additional syncs were 
determined to provide no additional durability guarantees and have been 
removed. See [ACCUMULO-4112][ACCUMULO-4112] for additional detail.
@@ -40,13 +44,13 @@ A performance issue was identified and corrected 
([ACCUMULO-1755][ACCUMULO-1755]
 
 ## Other Notable Changes
 
- * [ACCUMULO-4146][ACCUMULO-4146] Avoid copy of RFile Index Blocks when 
already in cache.
- * [ACCUMULO-4335][ACCUMULO-4335] Error conditions that result in a Halt 
should ensure non-zero process exit code.
- * [ACCUMULO-4334][ACCUMULO-4334] Ingest rates reported through JMX did not 
match rates reported by Monitor.
- * [ACCUMULO-4314][ACCUMULO-4314] Use statistics to choose better keys for 
RFile index.
  * [ACCUMULO-3923][ACCUMULO-3923] bootstrap_hdfs.sh script would copy 
incorrect jars to hdfs.
+ * [ACCUMULO-4146][ACCUMULO-4146] Avoid copy of RFile Index Blocks when 
already in cache.
  * [ACCUMULO-4155][ACCUMULO-4155] No longer publish javadoc for non-public API 
to website. (Still available in javadoc jars in maven)
  * [ACCUMULO-4173][ACCUMULO-4173] Provide balancer to balance table within 
subset of hosts.
+ * [ACCUMULO-4314][ACCUMULO-4314] Use statistics to choose better keys for 
RFile index.
+ * [ACCUMULO-4334][ACCUMULO-4334] Ingest rates reported through JMX did not 
match rates reported by Monitor.
+ * [ACCUMULO-4335][ACCUMULO-4335] Error conditions that result in a Halt 
should ensure non-zero process exit code.
 
 ## Testing
 
@@ -78,3 +82,4 @@ HDFS High-Availability instances, forcing NameNode failover.
 [ACCUMULO-4155]: https://issues.apache.org/jira/browse/ACCUMULO-4155
 [ACCUMULO-4173]: https://issues.apache.org/jira/browse/ACCUMULO-4173
 [ACCUMULO-4151]: https://issues.apache.org/jira/browse/ACCUMULO-4151
+[ACCUMULO-4164]: https://issues.apache.org/jira/browse/ACCUMULO-4164

Reply via email to