jamesgolick-cassandra 0.8.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,45 @@
1
+ # -*- encoding: utf-8 -*-
2
+
3
+ Gem::Specification.new do |s|
4
+ s.name = %q{jamesgolick-cassandra}
5
+ s.version = "0.8.2"
6
+
7
+ s.required_rubygems_version = Gem::Requirement.new(">= 0.8") if s.respond_to? :required_rubygems_version=
8
+ s.authors = ["Evan Weaver, Ryan King"]
9
+ s.date = %q{2010-04-13}
10
+ s.default_executable = %q{cassandra_helper}
11
+ s.description = %q{A Ruby client for the Cassandra distributed database.}
12
+ s.email = %q{}
13
+ s.executables = ["cassandra_helper"]
14
+ s.extra_rdoc_files = ["CHANGELOG", "LICENSE", "README.rdoc", "bin/cassandra_helper", "lib/cassandra.rb", "lib/cassandra/array.rb", "lib/cassandra/cassandra.rb", "lib/cassandra/columns.rb", "lib/cassandra/comparable.rb", "lib/cassandra/constants.rb", "lib/cassandra/debug.rb", "lib/cassandra/helpers.rb", "lib/cassandra/long.rb", "lib/cassandra/mock.rb", "lib/cassandra/ordered_hash.rb", "lib/cassandra/protocol.rb", "lib/cassandra/time.rb"]
15
+ s.files = ["CHANGELOG", "LICENSE", "Manifest", "README.rdoc", "Rakefile", "bin/cassandra_helper", "conf/cassandra.in.sh", "conf/log4j.properties", "conf/storage-conf.xml", "lib/cassandra.rb", "lib/cassandra/array.rb", "lib/cassandra/cassandra.rb", "lib/cassandra/columns.rb", "lib/cassandra/comparable.rb", "lib/cassandra/constants.rb", "lib/cassandra/debug.rb", "lib/cassandra/helpers.rb", "lib/cassandra/long.rb", "lib/cassandra/mock.rb", "lib/cassandra/ordered_hash.rb", "lib/cassandra/protocol.rb", "lib/cassandra/time.rb", "test/cassandra_client_test.rb", "test/cassandra_mock_test.rb", "test/cassandra_test.rb", "test/comparable_types_test.rb", "test/ordered_hash_test.rb", "test/test_helper.rb", "vendor/gen-rb/cassandra.rb", "vendor/gen-rb/cassandra_constants.rb", "vendor/gen-rb/cassandra_types.rb", "cassandra.gemspec", "test/eventmachine_test.rb"]
16
+ s.homepage = %q{http://blog.evanweaver.com/files/doc/fauna/cassandra/}
17
+ s.rdoc_options = ["--line-numbers", "--inline-source", "--title", "Cassandra", "--main", "README.rdoc"]
18
+ s.require_paths = ["lib"]
19
+ s.rubyforge_project = %q{fauna}
20
+ s.rubygems_version = %q{1.3.6}
21
+ s.summary = %q{A Ruby client for the Cassandra distributed database.}
22
+ s.test_files = ["test/cassandra_client_test.rb", "test/cassandra_mock_test.rb", "test/cassandra_test.rb", "test/comparable_types_test.rb", "test/eventmachine_test.rb", "test/ordered_hash_test.rb", "test/test_helper.rb"]
23
+
24
+ if s.respond_to? :specification_version then
25
+ current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
26
+ s.specification_version = 3
27
+
28
+ if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
29
+ s.add_runtime_dependency(%q<thrift_client>, [">= 0.4.0"])
30
+ s.add_runtime_dependency(%q<json_pure>, [">= 0"])
31
+ s.add_runtime_dependency(%q<rake>, [">= 0"])
32
+ s.add_runtime_dependency(%q<simple_uuid>, [">= 0.1.0"])
33
+ else
34
+ s.add_dependency(%q<thrift_client>, [">= 0.4.0"])
35
+ s.add_dependency(%q<json>, [">= 0"])
36
+ s.add_dependency(%q<rake>, [">= 0"])
37
+ s.add_dependency(%q<simple_uuid>, [">= 0.1.0"])
38
+ end
39
+ else
40
+ s.add_dependency(%q<thrift_client>, [">= 0.4.0"])
41
+ s.add_dependency(%q<json>, [">= 0"])
42
+ s.add_dependency(%q<rake>, [">= 0"])
43
+ s.add_dependency(%q<simple_uuid>, [">= 0.1.0"])
44
+ end
45
+ end
@@ -0,0 +1,47 @@
1
+ # Licensed to the Apache Software Foundation (ASF) under one
2
+ # or more contributor license agreements. See the NOTICE file
3
+ # distributed with this work for additional information
4
+ # regarding copyright ownership. The ASF licenses this file
5
+ # to you under the Apache License, Version 2.0 (the
6
+ # "License"); you may not use this file except in compliance
7
+ # with the License. You may obtain a copy of the License at
8
+ #
9
+ # http://www.apache.org/licenses/LICENSE-2.0
10
+ #
11
+ # Unless required by applicable law or agreed to in writing, software
12
+ # distributed under the License is distributed on an "AS IS" BASIS,
13
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14
+ # See the License for the specific language governing permissions and
15
+ # limitations under the License.
16
+
17
+ # The directory where Cassandra's configs live (required)
18
+ CASSANDRA_CONF=$CASSANDRA_CONF
19
+
20
+ # This can be the path to a jar file, or a directory containing the
21
+ # compiled classes. NOTE: This isn't needed by the startup script,
22
+ # it's just used here in constructing the classpath.
23
+ cassandra_bin=$CASSANDRA_HOME/build/classes
24
+
25
+ # The java classpath (required)
26
+ CLASSPATH=$CASSANDRA_CONF:$CASSANDRA_BIN
27
+
28
+ for jar in $CASSANDRA_HOME/lib/*.jar $CASSANDRA_HOME/build/lib/jars/*.jar; do
29
+ CLASSPATH=$CLASSPATH:$jar
30
+ done
31
+
32
+ # Arguments to pass to the JVM
33
+ JVM_OPTS=" \
34
+ -ea \
35
+ -Xms128M \
36
+ -Xmx1G \
37
+ -XX:TargetSurvivorRatio=90 \
38
+ -XX:+AggressiveOpts \
39
+ -XX:+UseParNewGC \
40
+ -XX:+UseConcMarkSweepGC \
41
+ -XX:+CMSParallelRemarkEnabled \
42
+ -XX:+HeapDumpOnOutOfMemoryError \
43
+ -XX:SurvivorRatio=128 \
44
+ -XX:MaxTenuringThreshold=0 \
45
+ -Dcom.sun.management.jmxremote.port=8080 \
46
+ -Dcom.sun.management.jmxremote.ssl=false \
47
+ -Dcom.sun.management.jmxremote.authenticate=false"
@@ -0,0 +1,38 @@
1
+ # Licensed to the Apache Software Foundation (ASF) under one
2
+ # or more contributor license agreements. See the NOTICE file
3
+ # distributed with this work for additional information
4
+ # regarding copyright ownership. The ASF licenses this file
5
+ # to you under the Apache License, Version 2.0 (the
6
+ # "License"); you may not use this file except in compliance
7
+ # with the License. You may obtain a copy of the License at
8
+ #
9
+ # http://www.apache.org/licenses/LICENSE-2.0
10
+ #
11
+ # Unless required by applicable law or agreed to in writing, software
12
+ # distributed under the License is distributed on an "AS IS" BASIS,
13
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14
+ # See the License for the specific language governing permissions and
15
+ # limitations under the License.
16
+
17
+ # for production, you should probably set the root to INFO
18
+ # and the pattern to %c instead of %l. (%l is slower.)
19
+
20
+ # output messages into a rolling log file as well as stdout
21
+ log4j.rootLogger=DEBUG,stdout,R
22
+
23
+ # stdout
24
+ log4j.appender.stdout=org.apache.log4j.ConsoleAppender
25
+ log4j.appender.stdout.layout=org.apache.log4j.SimpleLayout
26
+
27
+ # rolling log file ("system.log
28
+ log4j.appender.R=org.apache.log4j.DailyRollingFileAppender
29
+ log4j.appender.R.DatePattern='.'yyyy-MM-dd-HH
30
+ log4j.appender.R.layout=org.apache.log4j.PatternLayout
31
+ log4j.appender.R.layout.ConversionPattern=%5p [%t] %d{ISO8601} %F (line %L) %m%n
32
+ # Edit the next line to point to your logs directory
33
+ log4j.appender.R.File=data/logs/system.log
34
+
35
+ # Application logging options
36
+ #log4j.logger.com.facebook=DEBUG
37
+ #log4j.logger.com.facebook.infrastructure.gms=DEBUG
38
+ #log4j.logger.com.facebook.infrastructure.db=DEBUG
@@ -0,0 +1,340 @@
1
+ <!--
2
+ ~ Licensed to the Apache Software Foundation (ASF) under one
3
+ ~ or more contributor license agreements. See the NOTICE file
4
+ ~ distributed with this work for additional information
5
+ ~ regarding copyright ownership. The ASF licenses this file
6
+ ~ to you under the Apache License, Version 2.0 (the
7
+ ~ "License"); you may not use this file except in compliance
8
+ ~ with the License. You may obtain a copy of the License at
9
+ ~
10
+ ~ http://www.apache.org/licenses/LICENSE-2.0
11
+ ~
12
+ ~ Unless required by applicable law or agreed to in writing,
13
+ ~ software distributed under the License is distributed on an
14
+ ~ "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
15
+ ~ KIND, either express or implied. See the License for the
16
+ ~ specific language governing permissions and limitations
17
+ ~ under the License.
18
+ -->
19
+ <Storage>
20
+ <!--======================================================================-->
21
+ <!-- Basic Configuration -->
22
+ <!--======================================================================-->
23
+
24
+ <!--
25
+ ~ The name of this cluster. This is mainly used to prevent machines in
26
+ ~ one logical cluster from joining another.
27
+ -->
28
+ <ClusterName>Test</ClusterName>
29
+
30
+ <!--
31
+ ~ Turn on to make new [non-seed] nodes automatically migrate the right data
32
+ ~ to themselves. (If no InitialToken is specified, they will pick one
33
+ ~ such that they will get half the range of the most-loaded node.)
34
+ ~ If a node starts up without bootstrapping, it will mark itself bootstrapped
35
+ ~ so that you can't subsequently accidently bootstrap a node with
36
+ ~ data on it. (You can reset this by wiping your data and commitlog
37
+ ~ directories.)
38
+ ~
39
+ ~ Off by default so that new clusters and upgraders from 0.4 don't
40
+ ~ bootstrap immediately. You should turn this on when you start adding
41
+ ~ new nodes to a cluster that already has data on it. (If you are upgrading
42
+ ~ from 0.4, start your cluster with it off once before changing it to true.
43
+ ~ Otherwise, no data will be lost but you will incur a lot of unnecessary
44
+ ~ I/O before your cluster starts up.)
45
+ -->
46
+ <AutoBootstrap>false</AutoBootstrap>
47
+
48
+ <!--
49
+ ~ Keyspaces and ColumnFamilies:
50
+ ~ A ColumnFamily is the Cassandra concept closest to a relational
51
+ ~ table. Keyspaces are separate groups of ColumnFamilies. Except in
52
+ ~ very unusual circumstances you will have one Keyspace per application.
53
+
54
+ ~ There is an implicit keyspace named 'system' for Cassandra internals.
55
+ -->
56
+ <Keyspaces>
57
+ <Keyspace Name="Twitter">
58
+ <KeysCachedFraction>0.01</KeysCachedFraction>
59
+ <ColumnFamily CompareWith="UTF8Type" Name="Users" />
60
+ <ColumnFamily CompareWith="UTF8Type" Name="UserAudits" />
61
+ <ColumnFamily CompareWith="UTF8Type" CompareSubcolumnsWith="TimeUUIDType" ColumnType="Super" Name="UserRelationships" />
62
+ <ColumnFamily CompareWith="UTF8Type" Name="Usernames" />
63
+ <ColumnFamily CompareWith="UTF8Type" Name="Statuses" />
64
+ <ColumnFamily CompareWith="UTF8Type" Name="StatusAudits" />
65
+ <ColumnFamily CompareWith="UTF8Type" CompareSubcolumnsWith="TimeUUIDType" ColumnType="Super" Name="StatusRelationships" />
66
+ <ColumnFamily CompareWith="UTF8Type" ColumnType="Super" Name="Index" />
67
+ <ColumnFamily CompareWith="BytesType" ColumnType="Standard" Name="TimelinishThings" />
68
+
69
+ <ReplicaPlacementStrategy>org.apache.cassandra.locator.RackUnawareStrategy</ReplicaPlacementStrategy>
70
+ <ReplicationFactor>1</ReplicationFactor>
71
+ <EndPointSnitch>org.apache.cassandra.locator.EndPointSnitch</EndPointSnitch>
72
+ </Keyspace>
73
+
74
+ <Keyspace Name="Multiblog">
75
+ <KeysCachedFraction>0.01</KeysCachedFraction>
76
+ <ColumnFamily CompareWith="TimeUUIDType" Name="Blogs"/>
77
+ <ColumnFamily CompareWith="TimeUUIDType" Name="Comments"/>
78
+
79
+ <ReplicaPlacementStrategy>org.apache.cassandra.locator.RackUnawareStrategy</ReplicaPlacementStrategy>
80
+ <ReplicationFactor>1</ReplicationFactor>
81
+ <EndPointSnitch>org.apache.cassandra.locator.EndPointSnitch</EndPointSnitch>
82
+ </Keyspace>
83
+
84
+ <Keyspace Name="MultiblogLong">
85
+ <KeysCachedFraction>0.01</KeysCachedFraction>
86
+ <ColumnFamily CompareWith="LongType" Name="Blogs"/>
87
+ <ColumnFamily CompareWith="LongType" Name="Comments"/>
88
+
89
+ <ReplicaPlacementStrategy>org.apache.cassandra.locator.RackUnawareStrategy</ReplicaPlacementStrategy>
90
+ <ReplicationFactor>1</ReplicationFactor>
91
+ <EndPointSnitch>org.apache.cassandra.locator.EndPointSnitch</EndPointSnitch>
92
+ </Keyspace>
93
+
94
+ <Keyspace Name="CassandraObject">
95
+ <KeysCachedFraction>0.01</KeysCachedFraction>
96
+ <ColumnFamily CompareWith="UTF8Type" Name="Customers" />
97
+ <ColumnFamily CompareWith="UTF8Type" CompareSubcolumnsWith="TimeUUIDType" ColumnType="Super" Name="CustomerRelationships" />
98
+ <ColumnFamily CompareWith="UTF8Type" CompareSubcolumnsWith="TimeUUIDType" ColumnType="Super" Name="CustomersByLastName" />
99
+ <ColumnFamily CompareWith="UTF8Type" Name="Invoices" />
100
+ <ColumnFamily CompareWith="UTF8Type" CompareSubcolumnsWith="TimeUUIDType" ColumnType="Super" Name="InvoiceRelationships" />
101
+ <ColumnFamily CompareWith="UTF8Type" Name="InvoicesByNumber" />
102
+ <ColumnFamily CompareWith="UTF8Type" Name="Payments" />
103
+ <ColumnFamily CompareWith="UTF8Type" Name="Appointments" />
104
+
105
+ <ReplicaPlacementStrategy>org.apache.cassandra.locator.RackUnawareStrategy</ReplicaPlacementStrategy>
106
+ <ReplicationFactor>1</ReplicationFactor>
107
+ <EndPointSnitch>org.apache.cassandra.locator.EndPointSnitch</EndPointSnitch>
108
+ </Keyspace>
109
+ </Keyspaces>
110
+
111
+ <!--
112
+ ~ Authenticator: any IAuthenticator may be used, including your own as long
113
+ ~ as it is on the classpath. Out of the box, Cassandra provides
114
+ ~ org.apache.cassandra.auth.AllowAllAuthenticator and,
115
+ ~ org.apache.cassandra.auth.SimpleAuthenticator
116
+ ~ (SimpleAuthenticator uses access.properties and passwd.properties by
117
+ ~ default).
118
+ ~
119
+ ~ If you don't specify an authenticator, AllowAllAuthenticator is used.
120
+ -->
121
+ <Authenticator>org.apache.cassandra.auth.AllowAllAuthenticator</Authenticator>
122
+
123
+ <!--
124
+ ~ Partitioner: any IPartitioner may be used, including your own as long
125
+ ~ as it is on the classpath. Out of the box, Cassandra provides
126
+ ~ org.apache.cassandra.dht.RandomPartitioner,
127
+ ~ org.apache.cassandra.dht.OrderPreservingPartitioner, and
128
+ ~ org.apache.cassandra.dht.CollatingOrderPreservingPartitioner.
129
+ ~ (CollatingOPP colates according to EN,US rules, not naive byte
130
+ ~ ordering. Use this as an example if you need locale-aware collation.)
131
+ ~ Range queries require using an order-preserving partitioner.
132
+ ~
133
+ ~ Achtung! Changing this parameter requires wiping your data
134
+ ~ directories, since the partitioner can modify the sstable on-disk
135
+ ~ format.
136
+ -->
137
+ <Partitioner>org.apache.cassandra.dht.RandomPartitioner</Partitioner>
138
+
139
+ <!--
140
+ ~ If you are using an order-preserving partitioner and you know your key
141
+ ~ distribution, you can specify the token for this node to use. (Keys
142
+ ~ are sent to the node with the "closest" token, so distributing your
143
+ ~ tokens equally along the key distribution space will spread keys
144
+ ~ evenly across your cluster.) This setting is only checked the first
145
+ ~ time a node is started.
146
+
147
+ ~ This can also be useful with RandomPartitioner to force equal spacing
148
+ ~ of tokens around the hash space, especially for clusters with a small
149
+ ~ number of nodes.
150
+ -->
151
+ <InitialToken></InitialToken>
152
+
153
+ <!--
154
+ ~ Directories: Specify where Cassandra should store different data on
155
+ ~ disk. Keep the data disks and the CommitLog disks separate for best
156
+ ~ performance
157
+ -->
158
+ <CommitLogDirectory>data/cassandra/commitlog</CommitLogDirectory>
159
+ <DataFileDirectories>
160
+ <DataFileDirectory>data/cassandra/data</DataFileDirectory>
161
+ </DataFileDirectories>
162
+ <CalloutLocation>data/cassandra/callouts</CalloutLocation>
163
+ <StagingFileDirectory>data/cassandra/staging</StagingFileDirectory>
164
+
165
+
166
+ <!--
167
+ ~ Addresses of hosts that are deemed contact points. Cassandra nodes
168
+ ~ use this list of hosts to find each other and learn the topology of
169
+ ~ the ring. You must change this if you are running multiple nodes!
170
+ -->
171
+ <Seeds>
172
+ <Seed>127.0.0.1</Seed>
173
+ </Seeds>
174
+
175
+
176
+ <!-- Miscellaneous -->
177
+
178
+ <!-- Time to wait for a reply from other nodes before failing the command -->
179
+ <RpcTimeoutInMillis>5000</RpcTimeoutInMillis>
180
+ <!-- Size to allow commitlog to grow to before creating a new segment -->
181
+ <CommitLogRotationThresholdInMB>128</CommitLogRotationThresholdInMB>
182
+
183
+
184
+ <!-- Local hosts and ports -->
185
+
186
+ <!--
187
+ ~ Address to bind to and tell other nodes to connect to. You _must_
188
+ ~ change this if you want multiple nodes to be able to communicate!
189
+ ~
190
+ ~ Leaving it blank leaves it up to InetAddress.getLocalHost(). This
191
+ ~ will always do the Right Thing *if* the node is properly configured
192
+ ~ (hostname, name resolution, etc), and the Right Thing is to use the
193
+ ~ address associated with the hostname (it might not be).
194
+ -->
195
+ <ListenAddress>localhost</ListenAddress>
196
+ <!-- internal communications port -->
197
+ <StoragePort>7000</StoragePort>
198
+
199
+ <!--
200
+ ~ The address to bind the Thrift RPC service to. Unlike ListenAddress
201
+ ~ above, you *can* specify 0.0.0.0 here if you want Thrift to listen on
202
+ ~ all interfaces.
203
+ ~
204
+ ~ Leaving this blank has the same effect it does for ListenAddress,
205
+ ~ (i.e. it will be based on the configured hostname of the node).
206
+ -->
207
+ <ThriftAddress>localhost</ThriftAddress>
208
+ <!-- Thrift RPC port (the port clients connect to). -->
209
+ <ThriftPort>9160</ThriftPort>
210
+ <!--
211
+ ~ Whether or not to use a framed transport for Thrift. If this option
212
+ ~ is set to true then you must also use a framed transport on the
213
+ ~ client-side, (framed and non-framed transports are not compatible).
214
+ -->
215
+ <ThriftFramedTransport>false</ThriftFramedTransport>
216
+
217
+
218
+ <!--======================================================================-->
219
+ <!-- Memory, Disk, and Performance -->
220
+ <!--======================================================================-->
221
+
222
+ <!--
223
+ ~ Access mode. mmapped i/o is substantially faster, but only practical on
224
+ ~ a 64bit machine (which notably does not include EC2 "small" instances)
225
+ ~ or relatively small datasets. "auto", the safe choice, will enable
226
+ ~ mmapping on a 64bit JVM. Other values are "mmap", "mmap_index_only"
227
+ ~ (which may allow you to get part of the benefits of mmap on a 32bit
228
+ ~ machine by mmapping only index files) and "standard".
229
+ ~ (The buffer size settings that follow only apply to standard,
230
+ ~ non-mmapped i/o.)
231
+ -->
232
+ <DiskAccessMode>auto</DiskAccessMode>
233
+
234
+ <!--
235
+ ~ Buffer size to use when performing contiguous column slices. Increase
236
+ ~ this to the size of the column slices you typically perform.
237
+ ~ (Name-based queries are performed with a buffer size of
238
+ ~ ColumnIndexSizeInKB.)
239
+ -->
240
+ <SlicedBufferSizeInKB>64</SlicedBufferSizeInKB>
241
+
242
+ <!--
243
+ ~ Buffer size to use when flushing memtables to disk. (Only one
244
+ ~ memtable is ever flushed at a time.) Increase (decrease) the index
245
+ ~ buffer size relative to the data buffer if you have few (many)
246
+ ~ columns per key. Bigger is only better _if_ your memtables get large
247
+ ~ enough to use the space. (Check in your data directory after your
248
+ ~ app has been running long enough.) -->
249
+ <FlushDataBufferSizeInMB>32</FlushDataBufferSizeInMB>
250
+ <FlushIndexBufferSizeInMB>8</FlushIndexBufferSizeInMB>
251
+
252
+ <!--
253
+ ~ Add column indexes to a row after its contents reach this size.
254
+ ~ Increase if your column values are large, or if you have a very large
255
+ ~ number of columns. The competing causes are, Cassandra has to
256
+ ~ deserialize this much of the row to read a single column, so you want
257
+ ~ it to be small - at least if you do many partial-row reads - but all
258
+ ~ the index data is read for each access, so you don't want to generate
259
+ ~ that wastefully either.
260
+ -->
261
+ <ColumnIndexSizeInKB>64</ColumnIndexSizeInKB>
262
+
263
+ <!--
264
+ ~ Flush memtable after this much data has been inserted, including
265
+ ~ overwritten data. There is one memtable per column family, and
266
+ ~ this threshold is based solely on the amount of data stored, not
267
+ ~ actual heap memory usage (there is some overhead in indexing the
268
+ ~ columns).
269
+ -->
270
+ <MemtableThroughputInMB>64</MemtableThroughputInMB>
271
+ <!--
272
+ ~ Throughput setting for Binary Memtables. Typically these are
273
+ ~ used for bulk load so you want them to be larger.
274
+ -->
275
+ <BinaryMemtableThroughputInMB>256</BinaryMemtableThroughputInMB>
276
+ <!--
277
+ ~ The maximum number of columns in millions to store in memory per
278
+ ~ ColumnFamily before flushing to disk. This is also a per-memtable
279
+ ~ setting. Use with MemtableThroughputInMB to tune memory usage.
280
+ -->
281
+ <MemtableOperationsInMillions>0.3</MemtableOperationsInMillions>
282
+ <!--
283
+ ~ The maximum time to leave a dirty memtable unflushed.
284
+ ~ (While any affected columnfamilies have unflushed data from a
285
+ ~ commit log segment, that segment cannot be deleted.)
286
+ ~ This needs to be large enough that it won't cause a flush storm
287
+ ~ of all your memtables flushing at once because none has hit
288
+ ~ the size or count thresholds yet. For production, a larger
289
+ ~ value such as 1440 is recommended.
290
+ -->
291
+ <MemtableFlushAfterMinutes>60</MemtableFlushAfterMinutes>
292
+
293
+ <!--
294
+ ~ Unlike most systems, in Cassandra writes are faster than reads, so
295
+ ~ you can afford more of those in parallel. A good rule of thumb is 2
296
+ ~ concurrent reads per processor core. Increase ConcurrentWrites to
297
+ ~ the number of clients writing at once if you enable CommitLogSync +
298
+ ~ CommitLogSyncDelay. -->
299
+ <ConcurrentReads>8</ConcurrentReads>
300
+ <ConcurrentWrites>32</ConcurrentWrites>
301
+
302
+ <!--
303
+ ~ CommitLogSync may be either "periodic" or "batch." When in batch
304
+ ~ mode, Cassandra won't ack writes until the commit log has been
305
+ ~ fsynced to disk. It will wait up to CommitLogSyncBatchWindowInMS
306
+ ~ milliseconds for other writes, before performing the sync.
307
+
308
+ ~ This is less necessary in Cassandra than in traditional databases
309
+ ~ since replication reduces the odds of losing data from a failure
310
+ ~ after writing the log entry but before it actually reaches the disk.
311
+ ~ So the other option is "timed," where writes may be acked immediately
312
+ ~ and the CommitLog is simply synced every CommitLogSyncPeriodInMS
313
+ ~ milliseconds.
314
+ -->
315
+ <CommitLogSync>periodic</CommitLogSync>
316
+ <!--
317
+ ~ Interval at which to perform syncs of the CommitLog in periodic mode.
318
+ ~ Usually the default of 10000ms is fine; increase it if your i/o
319
+ ~ load is such that syncs are taking excessively long times.
320
+ -->
321
+ <CommitLogSyncPeriodInMS>10000</CommitLogSyncPeriodInMS>
322
+ <!--
323
+ ~ Delay (in milliseconds) during which additional commit log entries
324
+ ~ may be written before fsync in batch mode. This will increase
325
+ ~ latency slightly, but can vastly improve throughput where there are
326
+ ~ many writers. Set to zero to disable (each entry will be synced
327
+ ~ individually). Reasonable values range from a minimal 0.1 to 10 or
328
+ ~ even more if throughput matters more than latency.
329
+ -->
330
+ <!-- <CommitLogSyncBatchWindowInMS>1</CommitLogSyncBatchWindowInMS> -->
331
+
332
+ <!--
333
+ ~ Time to wait before garbage-collection deletion markers. Set this to
334
+ ~ a large enough value that you are confident that the deletion marker
335
+ ~ will be propagated to all replicas by the time this many seconds has
336
+ ~ elapsed, even in the face of hardware failures. The default value is
337
+ ~ ten days.
338
+ -->
339
+ <GCGraceSeconds>864000</GCGraceSeconds>
340
+ </Storage>