Grid technology uses many networked computers to provide more processing power than would be possible on a single computer. Grids provide the ability to perform thousands of computations or transactions at once, and in doing so offer a great opportunity to the financial industry to improve their performance and, as a result, profitability.

Grids can offer financial institutions a way to utilize their computing resources to provide optimal processing of computations. Diverse areas such as pricing derivatives, risk analysis and wealth management all process hundreds of computations to provide results, and would benefit greatly from the technologies that Grid systems have to offer — so why aren't more financial organizations taking advantage of Grid?

That was the question posed in a recent web conference titled “Grid Computing in Financial Services — Toward True IT Agility.” On hand were experts from the computing and financial arenas, who met to discuss strategies for resolving, implementing and optimizing a Grid system within the financial enterprise.

Representatives from Microsoft, Digipede Technologies and UBS, suggested that several issues are preventing the implementation of Grids into the financial sector, specifically a lack of standards, a lack of reliability, a lack of demonstrated case studies and perhaps most fundamental of all, the slow adaptation of applications for distributed computing.

An overwhelming sentiment was that the true potential of an enterprise Grid has not yet been imagined and customers have been held back in implementing systems that could offer remarkable benefits to their business.
 
Present was Stevan Vidich, an industry architect for the financial services group at Microsoft. Having worked with key industry players on the adoption of Microsoft's technology in financial markets, Vidich provides his thoughts on aiding the seamless integration of Grid technology in the financial industry.

“Based on feedback in this field we've seen that customers want an on-demand utility to dispatch work reliably and receive results easily — in real time,” Vidich explains. One of the fundamental aspects of integrating Grid systems into financial institutions is for customers to have these user-friendly applications available to them. “Customers want to simply chose an entry from a dropdown menu on Excel instructing, 'run on grid' and get results back in a couple of minutes.”

Vidich also explains that customers want support from mainstream vendors: “They want to be able to modify existing Windows applications, which can be improved by running on a two hundred node grid.”

“The limiting factors are our current ability to integrate these different components into enterprise. It's a huge challenge,” agrees Eric Kristoff. As  director of IT at UBS, Kristoff has spent much of his career advancing emerging technologies and distributed systems. He has supported and forged the UBS Grid computing strategy and has had experience in Grid usage in the finance industry.

Vidich takes this opportunity to explain how Microsoft Office 2007, with its significant performance enhancements, will be directly beneficial for this aspect of integration. Released earlier this year, Microsoft Office 2007 has made significant improvements to Excel, which, he says, “is a critical application for many organizations.” When it comes to high performance computing, Excel Service will be very relevant for Grid adoption in the financial sector. “In terms of high performance computing features, this version of Excel has several key advancements,” promises Vidich.

This new version of Excel has increased usability of spreadsheets, fundamental to financial institutions, where many users work on huge amounts of data. “Excel Services enables you to publish a spreadsheet onto a distributed network, allowing users to interact with the spreadsheet in various ways.” The spreadsheet is rendered for the end user with remarkable fidelity, and as it is browser based, does not demand the presence of Excel on a desktop. “The spreadsheet that gets displayed is like the real thing, the best part is that the designer who publishes the spreadsheet, now gets to chose which part of the data the end user can see, so they can limit what books and charts are available to whom,” says Vidich. These improvements will aid the adoption of Grid technologies into financial institutions by providing use of applications familiar to the majority of workers.

John Powers, founder and president of Digipede Technologies highlights why these improvements have been implemented so slowly — what has held Grid employment back? “It has taken us a decade to go from no grid to a few grids, to a handful of large grids.” He says it is not a hardware problem, and agrees with Vidich, “our biggest problem is adapting applications to the Grid.”

Complimenting Microsoft on their enhanced, 'Grid-friendly' version of Excel, Powers explains that most Grid solution products focus on Linux and can't handle the majority of applications that run on Windows. “We look at applications like Excel as fundamental to our services,” he states. “Let us make sure we can deal with Windows, that's where most of the skills in your organizations are. If we do this then we open up the power of Grid computing to a much bigger class of applications.”

Powers says focusing attention on smaller 'real world' grids will be more prosperous than looking at the bigger Grid systems: “We asked Grid computing customers what was wrong with their products and they were very clear: Grids need to scale up on real world applications. Its important to talk about 1000-node Grids but even more important to talk about 10-node Grids. These need to be easier to buy and use.”

With Microsoft improving common applications and tuning them into the needs of the Grid community, together with a new appreciation for the needs of the individual user, no doubt working examples of Grid technology in the financial services will appear in the near future, aiding the profitable adoption of Grids in Finance.