Permalink
Switch branches/tags
Nothing to show
Find file
Fetching contributors…
Cannot retrieve contributors at this time
28 lines (16 sloc) 921 Bytes
PREREQUISITES:
- Cassandra 1.1
- Hadoop 1.0.1+
- JDK 1.6
- Scala 2.9 (if building the Scala job)
1. Create the Cassandra schema with the following command:
cassandra-cli -h <cass_host> -f cassandra/cassandra_schema.txt
2. Use cassandra/seed_data.txt to populate some sample data (using the same command as above), or add your own data.
3. Increase the max heap size for the Hadoop client in hadoop-env.sh:
export HADOOP_CLIENT_OPTS="-Xmx1g $HADOOP_CLIENT_OPTS"
4. Copy jars from lib directory into $HADOOP_HOME/lib or otherwise ensure they are on the Hadoop classpath
5. cd into either java or scala, and run the build script to generate the jar file.
6. Execute the run script as follows:
./run <cass_host> <num_reducers> <keyspace> <input_column_family> <output_column_family>
Example if running locally using the schema script above:
./run localhost 1 HadoopTest TestInput TestOutput