#Cassandra TPC-DS

Setup

To use this, you need to also clone the tpcds-kit GitHub repo, as we'll use that to generate the data. It is expected that this repo and the tpcds-kit repo are both cloned from the same directory, resulting in a tpcds-kit/ and cassandra-tpcds/ directory in the same directory. If you do not do that, you will need to update the Makefile appropriately.

STEPS:

1. clone this repo:

git clone 

2. clone the tpcds-kit repo:

git clone https://github.com/grahn/tpcds-kit.git

3. Build the tpcds-kit binaries

cd tpcds-kit/tools
make -f Makefile.suite

4. Build the Java loader

cd ../../cassandra-tpcds
make compile

5. Make the data. You will need to update the Makefile with the desired TPCDS_SCALE_FACTOR. It ships with the value 1, which is only for development purposes.

You will need to update the IP_ADDR in Makefile to be one of the endpoints for your Cassandra cluster.

make data

6. Create the keyspace and tables in Cassandra

make ddl

7. Load the data into Cassandra

make loadall

##Queries The queries are Shark queries taken from the impala-tpcds-kit GitHub repo: https://github.com/cloudera/impala-tpcds-kit/tree/master/queries-sql92-modified/queries/shark

They are queries 3, 7, 19, 27, 34, 42, 43, 46, 52, 53, 55, 59, 63, 65, 68, 73, 79, 89, 98, and ss_max.

A version which has been slightly modified to reference database names, not just bare table names, is in queries_full.

To use these queries, run

dse spark
use tpcds;
source queries/q19.sql;

You can then do whichever query you want from within the Shark shell.