Storing and Retrieving Financial Transactions in Cassandra

This project provides an example of how to store transactions in Cassandra and calculate a balance roll-up.

Creating the schema

There is a ddl.cql file which you can use to create the keyspace, associated tables, types and user defined aggregates and functions. Copy this file to one of your Cassandra nodes and run:

cqlsh -f ddl.cql

Note: The keyspace is defined using SimpleStrategy as it's designed to run on a single node. Change the definition appropriately to increase the replication factor and/or strategy (if you're running with a multi-dc) cluster

Importing sample data

A sample CSV file is included under resources which you can use. Copy this to a Cassandra node and run cqlsh and use the following COPY command:

COPY txns_demo.transactions (acct_id,transaction_ts,transaction_id,transaction_amount,transaction_type,description,ttl_date) FROM 'sample-import.csv' WITH HEADER = true ;

Building

The project is built using sbt. In order to compile locally run sbt:

sbt

Then

compile

Running

Spark Roll Up

To run the Spark roll up, from the sbt prompt type:

runMain com.pragmasol.demo.SparkRollUp "cassandra.host=192.168.56.101"

Where you can pass the parameter specifying the cassandra host you're trying to connect to. Note, you must surround this parameter with quotes (as shown) or the parameters will not be passed correctly and an error will occur

Also be aware that this spark job runs locally. If you want to run with distributed workers/executors then you'll need to deploy spark appropriately and change the configuration to take advantage.

Balance Retrieval

To run the balance retrieval:

runMain com.pragmasol.demo.BalanceRetriever 38976122 911241

Supplying the account number and sortcode as parameters

License

This code is open source software licensed under the Apache 2.0 License.