Reliance Commercial Finance began operations in early 2007 and initially deployed Oracle Application Server and Oracle Database 10g to support its FinnOne core lending application, which is used by staff to complete commercial and home lending transactions. At the time, the application was hosted on a Sun Fire V40Z server and Oracle Database was hosted on a Sun Fire V890 server running eight UltraSPARC CPUs and 16GB of RAM.

“We had just entered the market and planned to become one of the top three lending organizations in India by 2010,” said Ravulapaty. “To reach that goal, we needed to make sure we served our customers as quickly and efficiently as possible. This would involve upgrading infrastructure to improve the performance of our core lending applications, more easily analyze our financial and customer data, and provide high availability features to eliminate unnecessary system downtime, which affects customer service.”

In 2008, Reliance Commercial Finance purchased a second Sun Fire V890 server running eight UltraSPARC CPUs and 16GB of RAM. It used Oracle Real Application Clusters to create a two-node server environment to provide high availability and redundancy features at its headquarters in Mumbai, India. The aim was to ensure the reliability of its FinnOne and other lending systems, so that if these systems went down, the business would still be able to process customers’ loan applications, repayments, and other financial transactions.

In March 2011, Reliance Commercial Finance migrated its high performance data warehouse, which enables the company to store and analyze financial transactions onto Oracle Exadata Database Machine X2-2 Quarter Rack, which features Oracle Database 11g running the Oracle Linux operating system. The Oracle Exadata Database Machine has 21TB of storage capacity.
These transactions include the amount of unpaid debt per month and customer data, such as the amount and term of a customer’s loan, and the amount and frequency of their repayments.

The company also created a disaster recovery center in Bangalore by deploying a Sun Fire V490 server with four UltraSPARC CPUs and 16GB of RAM. It deployed Oracle Active Data Guard to provide automatic, online data replication between servers at its primary site in Mumbai and its Bangalore disaster recovery center. This ensured the organization could recover from a disaster, such as a fire or anything else that could impair availability at its primary site.

Reliance Commercial Finance is running its FinnOne core lending application on Exadata Database Machine X2-2 Quarter Rack, and can now process customer loans and other account information up to 40% faster than previously.

This ensures customer queries, loan applications, repayments, and other financial transactions are completed quickly, which improves customer service levels.

“The performance of Oracle Exadata Database Machine is phenomenal. One of our very critical transaction processes—which used to take eight hours—is now completed within an hour,” said Ravulapaty. “This ensures we can serve more customers, more efficiently.”

With Oracle Exadata, Reliance Commercial Finance can process more customer transactions per day. “Oracle Exadata eases the load on our Oracle Database by distributing transaction workloads easily,” said Ravulapaty. “This ensures we can easily process up to 25,000 customer transactions per day. Previously, we processed about 5,000 customer transactions per day.”

Oracle Real Application Clusters also provides automatic failover capabilities, which means that if one server node fails, the data processing load automatically shifts across to the other server node. The company can be sure that its FinnOne application—used by staff to manage customer loans—will still be available in the event of a server failure.

“This ensures we can continue to help our customers by processing and approving their applications efficiently so they have money to buy a house or car, or invest in property,” said Ravulapaty.

Reliance Commercial Finance’s data warehouse—deployed on Oracle Exadata Database Machine—includes online business analytics tools; a barcode-based file tracking system used for loan application tracking; and an inventory funding system used for customer acquisition and maintenance of vehicle loans.

 These systems work with its FinnOne core lending application to complete loan processing, and data from each system is loaded into the company’s data warehouse for further analysis.

 “Over the next two years, we are expecting our customer numbers to increase by four times to between three and four million, which is likely to double our transaction load. We believe Oracle Exadata Database Machine will handle the increase easily, so there’s a lot of room for growth. We are also expecting our data warehouse to grow to 15TB between now and 2013.”

Oracle WebLogic Server was implemented to improve the performance of Reliance Commercial Finance’s file tracking system, by migrating the application from open source servers to Oracle Exadata Database Machine.

“Multiple users can now access this system almost immediately and we don’t have to deal with system compatibility and support issues that hampered our ability to service customers as efficiently as possible,” said Ravulapaty.

Reliance Commercial Finance’s transaction load increases significantly towards the end of the month as financial staff prepare for end-of-month reporting.

“It previously took between seven and eight hours to collate all the loan transactions for the month and provide it to the finance team, so they could compile reports,” said Ravulapaty.

“With Oracle Exadata Database Machine, the takes one hour, which means that financial staff and senior managers have faster access to transaction and customer information that enables them to make more informed decisions, such as which types of customers to target for up-selling or cross-selling.”

Reliance Commercial Finance uses Oracle GoldenGate to capture any changes in transaction and customer data stored on a number of Sun Fire V890 servers and automatically move this data in real-time to an Oracle Exadata Database Machine, located at its primary data center in Mumbai.

“Without Oracle GoldenGate in place, we wouldn’t be able to provide an automatic failover mechanism because data is saved in different formats on the Sun Fire servers and Oracle Exadata Database Machines,” said Ravulapaty. “It simply would not be retrievable.

“Because Oracle GoldenGate replicates data in real-time, we know that our data is always up to date and our operations won’t be affected in the event of a server failure at our primary site.”

“Our staff can be responding to customer queries based on the latest information, and senior managers can make decisions based on accurate information.”