accumulo-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject [1/2] accumulo git commit: Wrote blog post about durability and performance
Date Wed, 02 Nov 2016 18:46:34 GMT
Repository: accumulo
Updated Branches:
  refs/heads/gh-pages b8fa8b9de -> e0447313a

Wrote blog post about durability and performance


Branch: refs/heads/gh-pages
Commit: b02a14133ff7b8c89ef00706978f1a697bb1de47
Parents: 5611d6d
Author: Keith Turner <>
Authored: Thu Oct 27 22:55:30 2016 -0400
Committer: Keith Turner <>
Committed: Wed Nov 2 14:41:33 2016 -0400

 1.6/accumulo_user_manual.html                   |  12 ++
 _layouts/post.html                              |   7 +-
 .../blog/   | 174 +++++++++++++++++++
 3 files changed, 192 insertions(+), 1 deletion(-)
diff --git a/1.6/accumulo_user_manual.html b/1.6/accumulo_user_manual.html
index 7834ec8..509b673 100644
--- a/1.6/accumulo_user_manual.html
+++ b/1.6/accumulo_user_manual.html
@@ -762,6 +762,7 @@ span.icon > [class^="icon-"], span.icon > [class*=" icon-"] { cursor:
default; }
 <li><a href="#_tserver_wal_blocksize">tserver.wal.blocksize</a></li>
 <li><a href="#_tserver_wal_replication">tserver.wal.replication</a></li>
 <li><a href="#_tserver_wal_sync">tserver.wal.sync</a></li>
+<li><a href="#_tserver_wal_sync_method">tserver.wal.sync.method</a></li>
 <li><a href="#_tserver_walog_max_size">tserver.walog.max.size</a></li>
 <li><a href="#_tserver_walog_max_age">tserver.walog.max.age</a></li>
 <li><a href="#_tserver_workq_threads">tserver.workq.threads</a></li>
@@ -7100,6 +7101,17 @@ $HADOOP_PREFIX/share/hadoop/mapreduce/.*.jar,
 <div class="sect4">
+<h5 id="_tserver_wal_sync_method">tserver.wal.sync.method</h5>
+<div class="paragraph">
+<p>The method to invoke when sync'ing WALs. HSync will provide resiliency in the face
of unexpected power outages, at the cost of speed. If method is not available, the legacy
'sync' method will be used to ensure backwards compatibility with older Hadoop versions. A
value of 'hflush' is the alternative to the default value of 'hsync' which will result in
faster writes, but with less durability.  This property was added in Accumulo 1.6.1.</p>
+<div class="paragraph">
+<p><em>Type:</em> STRING<br>
+<em>Zookeeper Mutable:</em> yes<br>
+<em>Default Value:</em> <code>hsync</code></p>
+<div class="sect4">
 <h5 id="_tserver_walog_max_size">tserver.walog.max.size</h5>
 <div class="paragraph">
 <p>The maximum size for each write-ahead log. See comment for property tserver.memory.maps.max</p>
diff --git a/_layouts/post.html b/_layouts/post.html
index 5a9c338..5d42ad7 100644
--- a/_layouts/post.html
+++ b/_layouts/post.html
@@ -1,7 +1,12 @@
 layout: skeleton
-<h5>{% if %}{{ }} - {% endif %}{{ | date_to_string
}} </h5>
+{% if %}<tr><td><b>Author</b></td><td>&nbsp;&nbsp;{{ }} </td></tr>{% endif %}
+<tr><td><b>Post date </b></td><td> &nbsp;&nbsp;{{ | date_to_string }} </td></tr>
+{% if page.reviewers %}<tr><td><b>Reviewer(s) </b></td><td>
&nbsp;&nbsp;{{ page.reviewers }} </td></tr>{% endif %}
 {{ content }}
diff --git a/_posts/blog/ b/_posts/blog/
new file mode 100644
index 0000000..d079baa
--- /dev/null
+++ b/_posts/blog/
@@ -0,0 +1,174 @@
+title: "Durability Performance Implications"
+date: 2016-11-02 17:00:00 +0000
+author: Keith Turner
+reviewers: Josh Elser, Dave Marion, Christopher Tubbs
+## Overview
+Accumulo stores recently written data in a sorted in memory map.  Before data is
+added to this map, it's written to an unsorted write ahead log(WAL).  In the
+case when a tablet server dies, the recently written data is recovered from the
+When data is written to Accumulo the following happens :
+ * Client sends a batch of mutations to a tablet server
+ * Tablet server does the following :
+   * Writes mutation to tablet servers WAL
+   * Sync or flush tablet servers' WAL
+   * Adds mutations to sorted in memory map of each tablet.
+   * Reports success back to client.
+The sync/flush step above moves data written to the WAL from memory to disk.
+Write ahead logs are stored in HDFS. HDFS supports two ways of forcing data to
+disk for an open file : `hsync` and `hflush`.  
+## HDFS Sync/Flush Details
+When `hflush` is called on a WAL, it does not guarantee data is on disk.  It
+only guarantees that data is in OS buffers on each datanode and on its way to disk.
+As a result calls to `hflush` are very fast.  If a WAL is replicated to 3 data
+nodes then data may be lost if all three machines reboot or die.  If the datanode
+process dies, then data loss will not happen because the data was in OS buffers
+waiting to be written to disk.  The machines have to reboot or die for data loss to
+In order to avoid data loss in the event of reboot, `hsync` can be called.  This
+will ensure data is written to disk on all datanodes before returning.  When
+using `hsync` for the WAL, if Accumulo reports success to a user it means the
+data is on disk.  However `hsync` is much slower than `hflush` and the way it's
+implemented exacerbates the problem.  For example `hflush` make take 1ms and
+`hsync` may take 50ms.  This difference will impact writes to Accumulo and can
+be mitigated in some situations with larger buffers in Accumulo.
+HDFS keeps checksum data internally by default.  Datanodes store checksum data
+in a separate file in the local filesystem.  This means when `hsync` is called
+on a WAL, two files must be synced on each datanode.  Syncing two files doubles
+the time. To make matters even worse, when the two files are synced the local
+filesystem metadata is also synced.  Depending on the local filesystem and its
+configuration, syncing the metadata may or may not take time.  In the worst
+case, we need to wait for four sync operations at the local filesystem level on
+each datanode. One thing I am not sure about, is if these sync operations occur
+in parallel on the replicas on different datanodes.  If anyone can answer this
+question, please let us know on the [dev list][ML]. The following pointers show
+where sync occurs in the datanode code.
+ * [BlockReceiver.flushOrSync()][fos] calls [ReplicaOutputStreams.syncDataOut()][ros1] and
[ReplicaOutputStreams.syncChecksumOut()][ros2] when `isSync` is true.
+ * The methods in ReplicaOutputStreams call [FileChannel.force(true)][fcf] which
+   synchronously flushes data and filesystem metadata.
+If files were preallocated (this would avoid syncing local filesystem metadata)
+and checksums were stored in-line, then 1 sync could be done instead of 4.  
+## Configuring WAL flush/sync in Accumulo 1.6
+Accumulo 1.6.0 only supported `hsync` and this caused [performance
+problems][160_RN_WAL].  In order to offer better performance, the option to
+configure `hflush` was [added in 1.6.1][161_RN_WAL].  The
+[tserver.wal.sync.method][16_UM_SM] configuration option was added to support
+this feature.  This was a tablet server wide option that applied to everything
+written to any table.
+## Group Commit
+Each Accumulo tablet server has a single WAL.  When multiple clients send
+mutations to a tablet server at around the same time, the tablet sever may group
+all of this into a single WAL operation.  It will do this instead of writing and
+syncing or flushing each client's mutations to the WAL separately.  Doing this
+increase throughput and lowers average latency for clients.
+## Configuring WAL flush/sync in Accumulo 1.7+
+Accumulo 1.7.0 introduced [table.durability][17_UM_TD], a new per table property
+for configuring durability.  It also stopped using the `tserver.wal.sync.method`
+property.  The `table.durability` property has the following four legal values.
+This property defaults to the most durable option which is `sync`.
+ * **none** : Do not write to WAL            
+ * **log**  : Write to WAL, but do not sync  
+ * **flush** : Write to WAL and call `hflush` 
+ * **sync** : Write to WAL and call `hsync`  
+If multiple writes arrive at around the same time with different durability
+settings, then the group commit code will choose the most durable.  This can
+cause one tables settings to slow down writes to another table.  Basically, one
+table that is set to `sync` can impact the entire system.
+In Accumulo 1.6, it was easy to make all writes use `hflush` because there was
+only one tserver setting.  Getting everything to use `flush` in 1.7 and later
+can be a little tricky because by default the Accumulo metadata table is set to
+use `sync`.  The following shell commands show this. The first command sets
+`table.durability=flush` as a system wide default for all tables.  However, the
+metadata table is still set to `sync`, because it has a per table override for
+that setting.  This override is set when Accumulo is initialized.  To get this
+table to use `flush`, the per table override must be deleted.  After deleting
+those properties, the metadata tables will inherit the system wide setting.
+root@uno> config -s table.durability=flush
+root@uno> createtable foo
+root@uno foo> config -t foo -f table.durability
+SCOPE      | NAME                | VALUE
+default    | table.durability .. | sync
+system     |    @override ...... | flush
+root@uno> config -t accumulo.metadata -f table.durability
+SCOPE      | NAME                | VALUE
+default    | table.durability .. | sync
+system     |    @override ...... | flush
+table      |    @override ...... | sync
+root@uno> config -t accumulo.metadata -d table.durability
+root@uno> config -t accumulo.metadata -f table.durability
+SCOPE      | NAME                | VALUE
+default    | table.durability .. | sync
+system     |    @override ...... | flush
+In short, executing the following commands will make all writes use `flush`
+(assuming no other tables or namespaces have been specifically set to `sync`).
+config -s table.durability=flush
+config -t accumulo.metadata -d table.durability
+config -t accumulo.root -d table.durability
+Even with these settings adjusted, minor compactions could still force `hsync`
+to be called in 1.7.0 and 1.7.1.  This was fixed in 1.7.2 and 1.8.0.  See the
+[1.7.2 release notes][172_RN_MCHS] and [ACCUMULO-4112] for more details.
+In addition to the per table durability setting, a per batch writer durability
+setting was also added in 1.7.0.  See
+[BatchWriterConfig.setDurability(...)][SD].  This means any client could
+potentially cause a `hsync` operation to occur, even if the system is
+configured to use `hflush`.
+## Improving the situation
+The more granular durability settings introduced in 1.7.0 can cause some
+unexpected problems.  [ACCUMULO-4146] suggests one possible way to solve these
+problems with Per-durability write ahead logs.
+[160_RN_WAL]: {{ site.baseurl }}/release_notes/1.6.0#slower-writes-than-previous-accumulo-versions
+[161_RN_WAL]: {{ site.baseurl }}/release_notes/1.6.1#write-ahead-log-sync-implementation
+[16_UM_SM]: {{ site.baseurl }}/1.6/accumulo_user_manual#_tserver_wal_sync_method
+[17_UM_TD]: {{ site.baseurl }}/1.7/accumulo_user_manual#_table_durability
+[172_RN_MCHS]: {{ site.baseurl }}/release_notes/1.7.2#minor-performance-improvements
+[SD]: {{ site.baseurl }}/1.8/apidocs/org/apache/accumulo/core/client/BatchWriterConfig.html#setDurability(org.apache.accumulo.core.client.Durability)
+[ML]: {{ site.baseurl }}/mailing_list

View raw message