Systems/Enterprise:

SUN, QUEEN'S UNIV, FIRST DERIVATIVES TO SPEED BANK RISK ANALYSIS

A Grid computing-based service designed by the Belfast e-science center (BeSC) at Queen's University in conjunction with First Derivatives and powered by Sun Microsystems is dramatically reducing the time required to carry out risk assessment calculations for investment banks worldwide.

The new service has been designed to help financial institutions carry out a comprehensive risk assessment known as Value-at-Risk (VaR) on share portfolios using stock market transactions as input to detailed and complex simulations. VaR is a statistical measure, initially developed by J.P. Morgan in the 1980s for its own internal company-wide value-at-risk system.

Subsequently, in the 1990s VaR measures were combined with Profit and Loss statements (P&Ls) in a daily report to the Treasury in New York. Such calculations are computationally intensive as large amounts of financial derivatives calculations are required.

The daily activity on the NYSE alone can create up to 2GB of market transactions. The increasing demand for better application performance and consistent reliability continues to outstrip an organization's supply of available computational resources.

"Using an traditional 8 processor-based machine, these calculations can take up to four hours for a typical derivatives portfolio of 10,000 trades," said Paul Donachy, commercial director of Belfast e-Science center. "Using monte-carlo based simulations, vast amounts of calculations are undertaken and the high throughput data-access required in the calculations frequently creates bottlenecks. Now, with the new Grid based 32-node blade system in place, the calculation time for the above portfolio is reduced to only 20 minutes and at a fraction of the cost."

Currently, companies in the financial sector depend heavily on such computationally intensive calculations to gain a competitive advantage. An improvement in VaR calculation times provides traders and risk managers with more accurate assessment of the potential risk in undertaking derivatives transactions. Such increased accuracy in risk assessment has a direct impact on traders' margins.

At present, and in order to meet computational demands, such tasks are largely preformed on-site using distributed or multi-processor based systems.

Grid computing can offer dramatic improvements in the processing of risk management information by "sweating" the organization's existing distributed and under-used infrastructure, to make processing power transparently and instantly available where and when it is needed most.

The services devised by BeSC and First Derivatives are based on the Open Grid Services Architecture (OSGA) model and represents an evolution towards a Grid architecture based on Web services concepts and technologies.

The e-Science center designed a Grid solution running on the Sun Fire B1600 Blade platform loaded with B100x x86 Blades in a Sun rack 900. The overall Grid environment provides a low cost, high performance scalable alternative to the use of distributed or multi-processor based systems.