New Case Study Shows How Financial Services Firms Can More Effectively Harness the Power of HPC
Bellevue, WA – April 27, 2009 – According to a new case study conducted by Lab49 using ScaleOut StateServer, financial services firms can overcome performance limitations and more effectively harness the power of high-performance computing (HPC) with ScaleOut StateServer® Grid Computing Edition (SOSS/GCE).
High-performance computing is a story of eliminating bottlenecks – circumventing the limitations of a single machine, managing network usage, preventing oversubscription to scarce resources. Applications in financial services need to quickly process large amounts of data, and they often encounter bottlenecks in accessing data. Moreover, harnessing the power of multi-core, multi-server compute grids remains a formidable challenge.
SOSS/GCE was designed to address these challenges. Its distributed cache stages data in-memory within the compute grid to avoid bottlenecks in accessing backing storage. SOSS/GCE also introduces a unique new feature called “parallel method invocation” (PMI), which provides two key benefits. PMI enables applications to reach maximum performance by letting the distributed cache schedule work on the compute nodes where the data already resides, eliminating unnecessary data motion. It also creates a “map/reduce” application framework that dramatically reduces development time while automatically harnessing the full power of a compute grid.
ScaleOut Software, a leading provider of distributed in-memory caching for server farms and compute grids, partnered with Lab49 to evaluate the actual benefits of using ScaleOut StateServer Grid Computing Edition in a typical, real-world, financial services application. This study implemented a back-testing framework for automated equity trading strategies using SOSS/GCE’s new PMI feature.
The results were significant. The case study shows how SOSS/GCE helped this application deliver linearly scalable performance and avoid data motion which otherwise would saturate the grid’s interconnect, delivering more than 5.7 times faster throughput than conventional approaches. The case study also reports how the use of PMI’s easy-to-use framework kept design time to an absolute minimum.
About ScaleOut Software, Inc.
ScaleOut Software develops software products that provide scalable, highly available caching for workload data in server farms and compute grids. It has offices in Bellevue Washington and Beaverton, Oregon. The company was founded by Dr. William L. Bain, whose previous company, Valence Research, developed and distributed Web load-balancing software that was acquired by Microsoft Corporation and is now called Network Load Balancing within the Windows Server operating system.
For more information, contact David Brinker by email or phone at 503-643-3422.
ScaleOut Software, Inc. 10900 NE 4th Street, Suite 2300, Bellevue, WA 98004, T: 503-643-3422.
#####