BNP Paribas is a leader in European banking and financial services, with a significant and growing presence in the United States and leading positions in Asia. Like many companies in the financial services sector, BNP Paribas relies daily upon compute-intensive, business-critical applications that require high-performance, scalable and reliable processing power. Capital markets thrive on complexity, and sell-side institutions such as BNP Paribas rely on high-volume, complex trades to generate profits.

With one the largest international banking network, BNP Paribas is present in 85 countries and has close to 100,000 employees. For BNP Paribas, it was essential to employ the technologies that brought about not just IT efficiencies, but also translated into the highest level business benefits.

For the structured credit group within BNP Paribas' Credit Derivatives Operations (CDO), the ability to master highly complex scenarios was the key to generating profits and delivering new structured products to market. In an increasingly competitive industry, maintaining high margins on new products necessitated increasing compute capacity to run highly complex pricing and risk models accurately and reliably.

BNP Paribas wanted to solve problems related to running these business-critical pricing and risk models. For example, key users also faced challenges related to unacceptable report turnaround times, including:

• Risk figures that are critical to traders were often unavailable at the start of the trading day when data was late or when processing problems occurred.

• Complex and compute-intensive applications such as pricing scenarios for structured derivatives took hours to run, leading to significant lag times from the time a sales representative contacted a customer to the time a quote could be delivered. This latency translated into missed opportunities.

• Batch-oriented applications “stovepiped by trade” resulted in unproductive allocation of system resources. While some areas of the computing environment were idle, other parts required additional capacity driven by complexity and volume growth.

The problem was that the relevant business-critical applications at BNP Paribas resided on siloed systems and, therefore, the only way to meet increasing demand for processing power was by replicating the systems and purchasing new and costly hardware. In order to guarantee that enough processing power was available during surges in demand, IT resources were over-provisioned to provide sufficient “headroom.” Such over-provisioning of resources, as well as new hardware purchases, leads to increased IT spend for support personnel, bloating total cost of ownership (TCO). More importantly, by solving the problem in this fashion, the existing infrastructure was still not able to scale to meet the current or future needs of the organization.

In March, the global structured credit group, due to its increasing business volume and growth, sought a solution to help meet a surge in demand for IT processing power. BNP Paribas decided to fully leverage its existing resources using DataSynapse's GridServer Virtual Enterprise Edition. BNP Paribas could use this infrastructure software to virtualize data components and business logic found in applications and distribute these services across available system resources as needed, eliminating IT constraints and processing bottlenecks.

The structured credit department at BNP Paribas UK began the Grid implementation in August, and the application went live in four to six weeks, delivering immediate benefits to the enterprise. Deployed across HP blades, the Grid deployment helped to improve the performance of BNP Paribas' CDO business and to maximize cross-platform resources. The network consisted of 150 dual CPU servers, which ran in hyper-threaded mode; post Grid implementation, this effectively translated into 600 logical engines running on the grid. In sum, BNP Paribas was able to increase utilization of its existing hardware, rapidly improve application response time and accelerate application time to market.

Significant and immediate application performance gains were achieved, from time to build and deploy to improvements in scalability and resiliency. For BNP Paribas, it was clear that taking the first steps toward a service-oriented operating environment through Grid computing saved time, money and improved computing efficiency.

“We've seen an increase in the performance and resilience of our trading applications, enabling us to trade more complex exotics at larger volumes, while speeding overnight batches,” commented Dipak Shah, head of structured credit IT at BNP Paribas UK. “Grid lets us maximize compute and process resources without major investments in new hardware or expansion of our existing data centers.”

Other benefits that BNP Paribas has experienced include:

• Increased Revenues: BNP Paribas expanded its capabilities and offerings and accelerated time to market for new complex derivatives.

• Performance Gains: Grid enabled the easy migration of batch applications towards Web-based service offerings, which serves as a framework for SOA.

• Improved Report Turnaround Time: The time to process compute-intensive, complex risk figures was cut in half, and allows overnight batches to run faster and more efficiently. The result: timely delivery of business-critical information to front-office credit derivatives operations.

• Increased Trading Volume: Grid increased BNP Paribas's ability to trade larger volumes and allowed key traders to have valuable, reliable and accurate risk figures throughout the trading day.

• Cost Reduction and Containment: BNP Paribas reduced operating and capital costs and ongoing cost avoidance.

Through shared and aggregated IT resources, based on business needs, BNP Paribas broke down IT silos, increased utilization of existing hardware and reduced its capital spend requirements, while providing a dramatically improved application performance and response time. BNP Paribas has reaped the benefits of Grid technology and created new revenue-generating opportunities.

As BNP Paribas moves toward a global business strategy, the implementation of Grid facilitates expansion of the company's product areas. As a result of the successful Grid implementation in London, the company plans to expand its Grid project to New York and Tokyo, as well as across multiple business lines including the credit and interest rate derivatives group. Grid supports BNP Paribas' global expansion plans by providing cross-asset class computations to deliver value across the enterprise.

“With Grid, we've generated new business opportunities, reduced cycle times, and enabled greater transaction capacity, all driving revenue growth,” Shah remarked. “Grid computing is making us money, period, and we're looking forward to deploying our grid solution across global business operations, replicating our success in the global marketplace.”

About Kelly Vizzini

As chief marketing officer at DataSynapse, Kelly Vizzini works to leverage the company's existing successes and domain expertise to build a brand identity that positions DataSynapse as the de facto standard in the U.S. and European markets for distributed computing solutions. Prior to her role at DataSynapse, Vizzini held marketing positions at several software companies including Prescient, Optum, Metasys and InfoSystems. She holds a bachelor's degree in journalism and communications from the University of South Carolina.