Beaverton, OR and Toronto, ON – 18 September 2006

New Event Driven Design Paradigm Delivers 98% Performance Increase to Power Grid Deployments

GemStone Systems, the leading provider of the Enterprise Data Fabric (EDF), today announced the availability of a benchmarking report with Platform Computing. Focused on event driven grid calculations in a risk scenario, the benchmark results demonstrate a 98% improvement in efficiency and 61 times increase in net benefit in end-to-end client response time, offering firms the opportunity to develop and deploy new grid models while reducing both time-to-market and operational expense.

Executed on Intel Xeon 64-bit dual-core processor based hardware, the benchmark report highlights the benefits to be gained by combining Platform Symphony’s compute grid with the GemFire Enterprise Data Fabric data grid in a grid computing environment. By leveraging the power of the underlying Intel hardware platform, the convergence of the Platform compute grid and GemStone data grid was shown to deliver a new event-driven design paradigm for calculating grid computations.

The event driven model triggers grid calculations upon data updates and hence the load on the compute grid is smoothed out, as opposed to the scenario of running all computations based on a client-request.

“The name of the game in grid is efficiency and bringing customers increased performance to positively affect the bottom line,” said Simon Lonsdale, Director of Strategic Alliances, Platform. “The combination of Symphony, GemFire and Intel platforms gives end users further performance benefits on the grid, providing additional value to customers who have employed Symphony to grow along with their business.”

The joint use of the compute and data grids offers additional unprecedented design patterns such as:

- Intelligent advanced positioning of static data in a distributed in-memory data grid for immediate access by the compute nodes
- Data aware routing that promotes data-compute maximization by enabling intelligently delivering tasks to the appropriate node based on the data requirements of a specific task
- Intermediate results in a workflow style grid process are cached and distributed, precluding the need for a staging database that is typically centralized, latent and not scalable under parallel loads

“In order to maximize grid infrastructures, firms must have the ability to provision the appropriate compute power while guaranteeing consistent, low-latency access to enterprise data,” noted Juan Menendez, SVP of Business Development, GemStone Systems. “The convergence of the Platform compute and GemStone data grid deployed on Intel hardware enables firms to confront the typical latency, performance and scalability challenges impeding many current grid deployments, opening up new revenue streams and leading to increased profitability.”

“Intel is committed to working with key software providers to introduce and drive innovative, advanced technology to the financial services market. Working closely with GemStone and Platform, the three companies offer a scaleable and functionality-rich high performance grid computing solution, all built on Intel’s high-performance dual-core Xeon 5100 processor,” said Eric Doyle, WW Director, Financial Services Solutions, Intel Corporation.

Become a bobsguide member to access the following

1. Unrestricted access to bobsguide
2. Send a proposal request
3. Insights delivered daily to your inbox
4. Career development