cassandra_model_spark 0.0.1.5
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/README.md +17 -0
- data/bin/cmodel-spark-build +7 -0
- data/bin/cmodel-spark-env.rb +11 -0
- data/bin/cmodel-spark-master +22 -0
- data/bin/cmodel-spark-run-master +4 -0
- data/bin/cmodel-spark-run-master.sh +8 -0
- data/bin/cmodel-spark-run-slave +4 -0
- data/bin/cmodel-spark-run-slave.sh +8 -0
- data/bin/cmodel-spark-slaves +22 -0
- data/ext/scala_helper/bin/load-spark-env.sh +63 -0
- data/ext/scala_helper/bin/spark-class +87 -0
- data/ext/scala_helper/build.sbt +62 -0
- data/ext/scala_helper/cassandra_helper.scala +23 -0
- data/ext/scala_helper/data_type_helper.scala +27 -0
- data/ext/scala_helper/marshal_loader.scala +204 -0
- data/ext/scala_helper/marshal_row_mapping.scala +85 -0
- data/ext/scala_helper/project/plugins.sbt +6 -0
- data/ext/scala_helper/sbin/spark-config.sh +30 -0
- data/ext/scala_helper/sbin/spark-daemon.sh +223 -0
- data/ext/scala_helper/schema_builder.scala +35 -0
- data/ext/scala_helper/worker.scala +13 -0
- data/lib/cassandra_model_spark/build.rb +24 -0
- data/lib/cassandra_model_spark/column_cast.rb +44 -0
- data/lib/cassandra_model_spark/connection_cache.rb +9 -0
- data/lib/cassandra_model_spark/data_frame.rb +374 -0
- data/lib/cassandra_model_spark/java_bridge.rb +91 -0
- data/lib/cassandra_model_spark/java_classes.rb +36 -0
- data/lib/cassandra_model_spark/launcher.rb +150 -0
- data/lib/cassandra_model_spark/query_builder.rb +37 -0
- data/lib/cassandra_model_spark/raw_connection.rb +47 -0
- data/lib/cassandra_model_spark/record.rb +18 -0
- data/lib/cassandra_model_spark/spark.rb +33 -0
- data/lib/cassandra_model_spark.rb +42 -0
- metadata +127 -0
@@ -0,0 +1,30 @@
|
|
1
|
+
#
|
2
|
+
# Licensed to the Apache Software Foundation (ASF) under one or more
|
3
|
+
# contributor license agreements. See the NOTICE file distributed with
|
4
|
+
# this work for additional information regarding copyright ownership.
|
5
|
+
# The ASF licenses this file to You under the Apache License, Version 2.0
|
6
|
+
# (the "License"); you may not use this file except in compliance with
|
7
|
+
# 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
|
+
|
18
|
+
# included in all the spark scripts with source command
|
19
|
+
# should not be executable directly
|
20
|
+
# also should not be passed any arguments, since we need original $*
|
21
|
+
|
22
|
+
# symlink and absolute path should rely on SPARK_HOME to resolve
|
23
|
+
if [ -z "${SPARK_HOME}" ]; then
|
24
|
+
export SPARK_HOME="$(cd "`dirname "$0"`"/..; pwd)"
|
25
|
+
fi
|
26
|
+
|
27
|
+
export SPARK_CONF_DIR="${SPARK_CONF_DIR:-"${SPARK_HOME}/conf"}"
|
28
|
+
# Add the PySpark classes to the PYTHONPATH:
|
29
|
+
export PYTHONPATH="${SPARK_HOME}/python:${PYTHONPATH}"
|
30
|
+
export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.9-src.zip:${PYTHONPATH}"
|
@@ -0,0 +1,223 @@
|
|
1
|
+
#!/usr/bin/env bash
|
2
|
+
|
3
|
+
#
|
4
|
+
# Licensed to the Apache Software Foundation (ASF) under one or more
|
5
|
+
# contributor license agreements. See the NOTICE file distributed with
|
6
|
+
# this work for additional information regarding copyright ownership.
|
7
|
+
# The ASF licenses this file to You under the Apache License, Version 2.0
|
8
|
+
# (the "License"); you may not use this file except in compliance with
|
9
|
+
# the License. You may obtain a copy of the License at
|
10
|
+
#
|
11
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
12
|
+
#
|
13
|
+
# Unless required by applicable law or agreed to in writing, software
|
14
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
15
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
16
|
+
# See the License for the specific language governing permissions and
|
17
|
+
# limitations under the License.
|
18
|
+
#
|
19
|
+
|
20
|
+
# Runs a Spark command as a daemon.
|
21
|
+
#
|
22
|
+
# Environment Variables
|
23
|
+
#
|
24
|
+
# SPARK_CONF_DIR Alternate conf dir. Default is ${SPARK_HOME}/conf.
|
25
|
+
# SPARK_LOG_DIR Where log files are stored. ${SPARK_HOME}/logs by default.
|
26
|
+
# SPARK_MASTER host:path where spark code should be rsync'd from
|
27
|
+
# SPARK_PID_DIR The pid files are stored. /tmp by default.
|
28
|
+
# SPARK_IDENT_STRING A string representing this instance of spark. $USER by default
|
29
|
+
# SPARK_NICENESS The scheduling priority for daemons. Defaults to 0.
|
30
|
+
##
|
31
|
+
|
32
|
+
usage="Usage: spark-daemon.sh [--config <conf-dir>] (start|stop|submit|status) <spark-command> <spark-instance-number> <args...>"
|
33
|
+
|
34
|
+
# if no args specified, show usage
|
35
|
+
if [ $# -le 1 ]; then
|
36
|
+
echo $usage
|
37
|
+
exit 1
|
38
|
+
fi
|
39
|
+
|
40
|
+
if [ -z "${SPARK_HOME}" ]; then
|
41
|
+
export SPARK_HOME="$(cd "`dirname "$0"`"/..; pwd)"
|
42
|
+
fi
|
43
|
+
|
44
|
+
. "${SPARK_HOME}/sbin/spark-config.sh"
|
45
|
+
|
46
|
+
# get arguments
|
47
|
+
|
48
|
+
# Check if --config is passed as an argument. It is an optional parameter.
|
49
|
+
# Exit if the argument is not a directory.
|
50
|
+
|
51
|
+
if [ "$1" == "--config" ]
|
52
|
+
then
|
53
|
+
shift
|
54
|
+
conf_dir="$1"
|
55
|
+
if [ ! -d "$conf_dir" ]
|
56
|
+
then
|
57
|
+
echo "ERROR : $conf_dir is not a directory"
|
58
|
+
echo $usage
|
59
|
+
exit 1
|
60
|
+
else
|
61
|
+
export SPARK_CONF_DIR="$conf_dir"
|
62
|
+
fi
|
63
|
+
shift
|
64
|
+
fi
|
65
|
+
|
66
|
+
option=$1
|
67
|
+
shift
|
68
|
+
command=$1
|
69
|
+
shift
|
70
|
+
instance=$1
|
71
|
+
shift
|
72
|
+
|
73
|
+
spark_rotate_log ()
|
74
|
+
{
|
75
|
+
log=$1;
|
76
|
+
num=5;
|
77
|
+
if [ -n "$2" ]; then
|
78
|
+
num=$2
|
79
|
+
fi
|
80
|
+
if [ -f "$log" ]; then # rotate logs
|
81
|
+
while [ $num -gt 1 ]; do
|
82
|
+
prev=`expr $num - 1`
|
83
|
+
[ -f "$log.$prev" ] && mv "$log.$prev" "$log.$num"
|
84
|
+
num=$prev
|
85
|
+
done
|
86
|
+
mv "$log" "$log.$num";
|
87
|
+
fi
|
88
|
+
}
|
89
|
+
|
90
|
+
. "${SPARK_HOME}/bin/load-spark-env.sh"
|
91
|
+
|
92
|
+
if [ "$SPARK_IDENT_STRING" = "" ]; then
|
93
|
+
export SPARK_IDENT_STRING="$USER"
|
94
|
+
fi
|
95
|
+
|
96
|
+
|
97
|
+
export SPARK_PRINT_LAUNCH_COMMAND="1"
|
98
|
+
|
99
|
+
# get log directory
|
100
|
+
if [ "$SPARK_LOG_DIR" = "" ]; then
|
101
|
+
export SPARK_LOG_DIR="${SPARK_HOME}/logs"
|
102
|
+
fi
|
103
|
+
mkdir -p "$SPARK_LOG_DIR"
|
104
|
+
touch "$SPARK_LOG_DIR"/.spark_test > /dev/null 2>&1
|
105
|
+
TEST_LOG_DIR=$?
|
106
|
+
if [ "${TEST_LOG_DIR}" = "0" ]; then
|
107
|
+
rm -f "$SPARK_LOG_DIR"/.spark_test
|
108
|
+
else
|
109
|
+
chown "$SPARK_IDENT_STRING" "$SPARK_LOG_DIR"
|
110
|
+
fi
|
111
|
+
|
112
|
+
if [ "$SPARK_PID_DIR" = "" ]; then
|
113
|
+
SPARK_PID_DIR=/tmp
|
114
|
+
fi
|
115
|
+
|
116
|
+
# some variables
|
117
|
+
log="$SPARK_LOG_DIR/spark-$SPARK_IDENT_STRING-$command-$instance-$HOSTNAME.out"
|
118
|
+
pid="$SPARK_PID_DIR/spark-$SPARK_IDENT_STRING-$command-$instance.pid"
|
119
|
+
|
120
|
+
# Set default scheduling priority
|
121
|
+
if [ "$SPARK_NICENESS" = "" ]; then
|
122
|
+
export SPARK_NICENESS=0
|
123
|
+
fi
|
124
|
+
|
125
|
+
run_command() {
|
126
|
+
mode="$1"
|
127
|
+
shift
|
128
|
+
|
129
|
+
mkdir -p "$SPARK_PID_DIR"
|
130
|
+
|
131
|
+
if [ -f "$pid" ]; then
|
132
|
+
TARGET_ID="$(cat "$pid")"
|
133
|
+
if [[ $(ps -p "$TARGET_ID" -o comm=) =~ "java" ]]; then
|
134
|
+
echo "$command running as process $TARGET_ID. Stop it first."
|
135
|
+
exit 1
|
136
|
+
fi
|
137
|
+
fi
|
138
|
+
|
139
|
+
if [ "$SPARK_MASTER" != "" ]; then
|
140
|
+
echo rsync from "$SPARK_MASTER"
|
141
|
+
rsync -a -e ssh --delete --exclude=.svn --exclude='logs/*' --exclude='contrib/hod/logs/*' "$SPARK_MASTER/" "${SPARK_HOME}"
|
142
|
+
fi
|
143
|
+
|
144
|
+
spark_rotate_log "$log"
|
145
|
+
echo "starting $command, logging to $log"
|
146
|
+
|
147
|
+
case "$mode" in
|
148
|
+
(class)
|
149
|
+
nohup nice -n "$SPARK_NICENESS" "${SPARK_HOME}"/bin/spark-class $command "$@" >> "$log" 2>&1 < /dev/null &
|
150
|
+
newpid="$!"
|
151
|
+
;;
|
152
|
+
|
153
|
+
(submit)
|
154
|
+
nohup nice -n "$SPARK_NICENESS" "${SPARK_HOME}"/bin/spark-submit --class $command "$@" >> "$log" 2>&1 < /dev/null &
|
155
|
+
newpid="$!"
|
156
|
+
;;
|
157
|
+
|
158
|
+
(*)
|
159
|
+
echo "unknown mode: $mode"
|
160
|
+
exit 1
|
161
|
+
;;
|
162
|
+
esac
|
163
|
+
|
164
|
+
echo "$newpid" > "$pid"
|
165
|
+
sleep 2
|
166
|
+
# Check if the process has died; in that case we'll tail the log so the user can see
|
167
|
+
if [[ ! $(ps -p "$newpid" -o comm=) =~ "java" ]]; then
|
168
|
+
echo "failed to launch $command:"
|
169
|
+
tail -2 "$log" | sed 's/^/ /'
|
170
|
+
echo "full log in $log"
|
171
|
+
fi
|
172
|
+
}
|
173
|
+
|
174
|
+
case $option in
|
175
|
+
|
176
|
+
(submit)
|
177
|
+
run_command submit "$@"
|
178
|
+
;;
|
179
|
+
|
180
|
+
(start)
|
181
|
+
run_command class "$@"
|
182
|
+
;;
|
183
|
+
|
184
|
+
(stop)
|
185
|
+
|
186
|
+
if [ -f $pid ]; then
|
187
|
+
TARGET_ID="$(cat "$pid")"
|
188
|
+
if [[ $(ps -p "$TARGET_ID" -o comm=) =~ "java" ]]; then
|
189
|
+
echo "stopping $command"
|
190
|
+
kill "$TARGET_ID" && rm -f "$pid"
|
191
|
+
else
|
192
|
+
echo "no $command to stop"
|
193
|
+
fi
|
194
|
+
else
|
195
|
+
echo "no $command to stop"
|
196
|
+
fi
|
197
|
+
;;
|
198
|
+
|
199
|
+
(status)
|
200
|
+
|
201
|
+
if [ -f $pid ]; then
|
202
|
+
TARGET_ID="$(cat "$pid")"
|
203
|
+
if [[ $(ps -p "$TARGET_ID" -o comm=) =~ "java" ]]; then
|
204
|
+
echo $command is running.
|
205
|
+
exit 0
|
206
|
+
else
|
207
|
+
echo $pid file is present but $command not running
|
208
|
+
exit 1
|
209
|
+
fi
|
210
|
+
else
|
211
|
+
echo $command not running.
|
212
|
+
exit 2
|
213
|
+
fi
|
214
|
+
;;
|
215
|
+
|
216
|
+
(*)
|
217
|
+
echo $usage
|
218
|
+
exit 1
|
219
|
+
;;
|
220
|
+
|
221
|
+
esac
|
222
|
+
|
223
|
+
|
@@ -0,0 +1,35 @@
|
|
1
|
+
package org.apache.spark.api.cassandra_model
|
2
|
+
|
3
|
+
import org.apache.spark.rdd._
|
4
|
+
import com.datastax.spark.connector._
|
5
|
+
import com.datastax.spark.connector.rdd._
|
6
|
+
import org.apache.spark.sql._
|
7
|
+
import org.apache.spark.sql.types._
|
8
|
+
|
9
|
+
class SchemaBuilder {
|
10
|
+
var fields = Array[StructField]()
|
11
|
+
|
12
|
+
def build = StructType(fields)
|
13
|
+
|
14
|
+
def cassandraRDDToRDD(rdd: RDD[CassandraRow]) = {
|
15
|
+
rdd.map(
|
16
|
+
p => Row.fromSeq(
|
17
|
+
p.columnValues.map{
|
18
|
+
p => p match {
|
19
|
+
case (d: java.util.Date) => new java.sql.Timestamp(d.getTime())
|
20
|
+
case _ => p
|
21
|
+
}
|
22
|
+
}
|
23
|
+
)
|
24
|
+
)
|
25
|
+
}
|
26
|
+
|
27
|
+
def createDataFrame(sqlc: SQLContext, rdd: RDD[CassandraRow]) = {
|
28
|
+
val new_rdd = cassandraRDDToRDD(rdd)
|
29
|
+
sqlc.createDataFrame(new_rdd, build)
|
30
|
+
}
|
31
|
+
|
32
|
+
def addColumn(name: String, sql_type: DataType) = {
|
33
|
+
fields :+= StructField(name, sql_type, true)
|
34
|
+
}
|
35
|
+
}
|
@@ -0,0 +1,13 @@
|
|
1
|
+
package org.apache.spark.deploy.worker
|
2
|
+
|
3
|
+
import org.apache.spark.SparkConf
|
4
|
+
|
5
|
+
object RubyWorkerStarter {
|
6
|
+
def startWorker(master_url: String, host: String, port: Int, web_ui_port: Int, conf: SparkConf) = {
|
7
|
+
val temp_conf = new SparkConf
|
8
|
+
val argv = Array(master_url)
|
9
|
+
val args = new WorkerArguments(argv, temp_conf)
|
10
|
+
|
11
|
+
Worker.startRpcEnvAndEndpoint(host, port, web_ui_port, args.cores, args.memory, args.masters, args.workDir, conf = conf)
|
12
|
+
}
|
13
|
+
}
|
@@ -0,0 +1,24 @@
|
|
1
|
+
require 'optparse'
|
2
|
+
require_relative 'spark'
|
3
|
+
|
4
|
+
options = {}
|
5
|
+
OptionParser.new do |opts|
|
6
|
+
opts.banner = 'Usage: build.rb [--only-ext]'
|
7
|
+
opts.on('-e', '--only-ext', 'Build only extension') do
|
8
|
+
options[:only_ext] = true
|
9
|
+
end
|
10
|
+
end.parse!
|
11
|
+
|
12
|
+
ASSEMBLY_PATH = '/ext/scala_helper'
|
13
|
+
|
14
|
+
Dir.chdir("#{CassandraModel::Spark.root}#{ASSEMBLY_PATH}") do
|
15
|
+
puts '=> building extension...'
|
16
|
+
cmd = 'sbt package'
|
17
|
+
cmd << ' assemblyPackageDependency' unless options[:only_ext]
|
18
|
+
system(ENV.to_hash.merge('TARGET_DIR' => CassandraModel::Spark.classpath), cmd)
|
19
|
+
%w(bin sbin).each do |path|
|
20
|
+
puts "=> copying #{path}/ to #{CassandraModel::Spark.home}/"
|
21
|
+
`cp -R #{CassandraModel::Spark.root}#{ASSEMBLY_PATH}/#{path}/ #{CassandraModel::Spark.home}/`
|
22
|
+
end
|
23
|
+
`touch #{CassandraModel::Spark.home}/RELEASE`
|
24
|
+
end
|
@@ -0,0 +1,44 @@
|
|
1
|
+
module CassandraModel
|
2
|
+
module Spark
|
3
|
+
class ColumnCast
|
4
|
+
include ThomasUtils::SymbolHelpers
|
5
|
+
|
6
|
+
attr_reader :key
|
7
|
+
|
8
|
+
def initialize(key, type)
|
9
|
+
@key = key
|
10
|
+
@type = type.to_s.upcase
|
11
|
+
end
|
12
|
+
|
13
|
+
def quote(quote)
|
14
|
+
quoted_key = if @key.respond_to?(:quote)
|
15
|
+
@key.quote(quote)
|
16
|
+
else
|
17
|
+
"#{quote}#{@key}#{quote}"
|
18
|
+
end
|
19
|
+
"CAST(#{quoted_key} AS #{@type})"
|
20
|
+
end
|
21
|
+
|
22
|
+
def new_key(key)
|
23
|
+
self.class.new(key, @type)
|
24
|
+
end
|
25
|
+
|
26
|
+
end
|
27
|
+
end
|
28
|
+
end
|
29
|
+
|
30
|
+
module ThomasUtils
|
31
|
+
class KeyChild
|
32
|
+
def cast_as(type)
|
33
|
+
CassandraModel::Spark::ColumnCast.new(self, type)
|
34
|
+
end
|
35
|
+
alias :* :cast_as
|
36
|
+
end
|
37
|
+
end
|
38
|
+
|
39
|
+
class Symbol
|
40
|
+
def cast_as(type)
|
41
|
+
CassandraModel::Spark::ColumnCast.new(self, type)
|
42
|
+
end
|
43
|
+
alias :* :cast_as
|
44
|
+
end
|