Author: jbellis
Date: Mon Sep 12 17:42:07 2011
New Revision: 1169840
URL: http://svn.apache.org/viewvc?rev=1169840&view=rev
Log:
update help to explain benefits of SerializingCacheProvider
Modified:
cassandra/branches/cassandra-1.0.0/src/resources/org/apache/cassandra/cli/CliHelp.yaml
Modified:
cassandra/branches/cassandra-1.0.0/src/resources/org/apache/cassandra/cli/CliHelp.yaml
URL:
http://svn.apache.org/viewvc/cassandra/branches/cassandra-1.0.0/src/resources/org/apache/cassandra/cli/CliHelp.yaml?rev=1169840&r1=1169839&r2=1169840&view=diff
==============================================================================
---
cassandra/branches/cassandra-1.0.0/src/resources/org/apache/cassandra/cli/CliHelp.yaml
(original)
+++
cassandra/branches/cassandra-1.0.0/src/resources/org/apache/cassandra/cli/CliHelp.yaml
Mon Sep 12 17:42:07 2011
@@ -546,22 +546,23 @@ commands:
follower replicas. Accepts the values true and false.
- row_cache_provider: The provider for the row cache to use for this
- column family. Defaults to ConcurrentLinkedHashCacheProvider. .
+ column family.
Supported values are:
- ConcurrentLinkedHashCacheProvider
- - SerializingCacheProvider
+ - SerializingCacheProvider (requires JNA)
It is also valid to specify the fully-qualified class name to a class
that implements org.apache.cassandra.cache.IRowCacheProvider.
- ConcurrentLinkedHashCacheProvider provides the same features as the
versions
- prior to Cassandra v0.8. Row data is cached using the Java JVM heap.
-
- SerializingCacheProvider serialises the contents of the row and
- stores the data off the JVM Heap. This may reduce the GC pressure.
- NOTE: This provider requires JNA.jar to be in the class path to
- enable native methods.
+ row_cache_provider defaults to ConcurrentLinkedHashCacheProvider,
+ but if you have JNA installed you should usually use
+ SerializingCacheProvider, which serialises the contents of the
+ row and stores it in native memory, i.e., off the JVM
+ Heap. Serialized rows take significantly less memory than
+ "live" rows in the JVM, so you can cache more rows in a given
+ memory footprint. And storing the cache off-heap means you
+ can use smaller heap sizes, reducing the impact of GC pauses.
- compression: Use compression for SSTable data files.
@@ -818,22 +819,23 @@ commands:
follower replicas. Accepts the values true and false.
- row_cache_provider: The provider for the row cache to use for this
- column family. Defaults to ConcurrentLinkedHashCacheProvider. .
+ column family.
Supported values are:
- ConcurrentLinkedHashCacheProvider
- - SerializingCacheProvider
+ - SerializingCacheProvider (requires JNA)
It is also valid to specify the fully-qualified class name to a class
that implements org.apache.cassandra.cache.IRowCacheProvider.
- ConcurrentLinkedHashCacheProvider provides the same features as the
versions
- prior to Cassandra v0.8. Row data is cached using the Java JVM heap.
-
- SerializingCacheProvider serialises the contents of the row and
- stores the data off the JVM Heap. This may reduce the GC pressure.
- NOTE: Thsi provider requires JNA.jar to be in the class path to
- enable native methods.
+ row_cache_provider defaults to ConcurrentLinkedHashCacheProvider,
+ but if you have JNA installed you should usually use
+ SerializingCacheProvider, which serialises the contents of the
+ row and stores it in native memory, i.e., off the JVM
+ Heap. Serialized rows take significantly less memory than
+ "live" rows in the JVM, so you can cache more rows in a given
+ memory footprint. And storing the cache off-heap means you
+ can use smaller heap sizes, reducing the impact of GC pauses.
- compression: Use compression for SSTable data files. Accepts the
values true and false.
