In today's economic climate, everyone is looking for ways to reduce cost and improve performance. The cloud seems like an obvious answer. So does Linux and open source software. But how do you get there? How do extend your capacity and resources to the cloud in a cost-effective, risk-free way? New business models like appliances offer a stepping stone. By using Linux as the platform of choice in the heterogeneous, service-driven data center, customers gain the ability to reduce cost, complexity and risk at the same time as they deliver improved services to their end users.

Justin Steinman talked about the next generation of Linux, the service-driven data center, and how both concepts are helping customers extend their data center to the cloud.

• Justin Steinman, Vice President, Solution & Product Marketing, Novell

We brought together a panel from Industry and Financial Services to debate the finer points of establishing a Financial Services Cloud. While it is clear that establishing a Private Cloud can provide many of the economic benefits of Cloud Computing for a large enterprise there will always be value in moving beyond your firewall and going off premise. We discuss the pros and cons of establishing a Financial Services Cloud - provided by a trusted third party - with differentiated Financial Services Cloud Services backed by strong SLAs, as an alternative to some of the Public Cloud offerings available today. In short is there a sector Cloud play here?

• Duncan Johnston-Watt, CEO, CloudSoft (Moderator)

• Linda Bernardi, Founder, StraTerra Partners

• Jeff Birnbaum, Managing Director, Chief Technology Architect & Global Head of Architecture and Engineering, Merrill Lynch

• David Crosbie, CTO, Leostream

• Larry Tabb, Founder and CEO, TABB Group

Low latency delivery of market data has hitherto only been possible by heavy investment in proprietary ticker plant and messaging products.

Today, the combination of open source developments, open standards, and a user-driven desire to avoid vendor lock-in, is driving an alternative approach to building these architectures, which underpin new trading opportunities and business growth.

• Richard Reichgut, Director of Financial Services, Collaborative Software Initiative (Moderator)

• Jeff Birnbaum, Managing Director, Chief Technology Architect & Global Head of Architecture and Engineering, Merrill Lynch

• Angelo Corsaro, Ph.D., Product Strategy and Marketing Manager, PrismTech

• Ramakrishna Karedla, Performance Architect, High Performance Computing, Financial Services, Intel

• Alan Paris, Executive Director, Ernst and Young

• Sumeet Shendrikar, Technical Account Manager, Financial Services, RTI

This session reviewed a number of recent innovations that enable affordable high performance data storage at lower costs. Topics covered included clustered file systems, novel approaches to tiered storage and life cycle management, disk spindle virtualization, and flash-based solid state disk.

• Jacob Farmer, CTO, Cambridge Computer

Database analysis has today become a core strategic competency for many enterprises. Knowledge derived from analytics have become the key drivers of corporate strategy.

Evidence of the importance of analytics can be found in the ever-growing mountains of data being collected by enterprises across many markets. For example, banks, hedge funds, credit card and insurance companies, and others in the Financial Services industry depend heavily on analytics to handle processes such as risk profiling, portfolio pricing, marketing campaign management, and fraud detection. Data volumes today are easily in the terabyte-scale, even for medium-size businesses. Analysis of terabyte-scale databases requires planning and careful choice of computing/storage systems to achieve reasonable performance.

In recent years, the legacy "big Unix box" solutions are being challenged by two competing alternatives: fully-integrated "Appliances" and large Linux clusters. This panel with discuss the challenges, the emerging Appliance vendors, and the trade-offs versus commodity clusters.

• Geno Valente, VP, Sales and Marketing, XtremeData (Moderator)

• John Bair, CTO, LaunchPoint

• Gary Bhattacharjee, Business Intelligence Services, Capital Markets, HP

• Gordon S. Linoff, Founder, Data Miners

• Anil Sharma, Executive Director, IDEAS Practice Areas, Morgan Stanley

New multi-core microprocessor architectures are enabling the next generation of highly scalable, low latency financial markets applications. This session looked at specific new functionality and how application architects can leverage it.

• Charles Milo, Member of Senior Staff, Enterprise Technical Specialist, Intel