ScaleOut StateServer Grid Computing Edition Introduces Automatic, In-Memory, Map/Reduce
New York, NY – September 22, 2008 – From the floor of the HPC on Wall Street Conference, ScaleOut Software, Inc. today announced general availability for its new ScaleOut StateServer Grid Computing Edition™ (GCE). Using this new edition, customers can now leverage their distributed cache to perform user-specified “map/reduce” computations across a powerful compute grid and obtain automatic parallel speedup. Data-parallel grid computations performed to analyze complex financial data, such as a large set of portfolios, benefit from GCE’s significantly reduced processing time and its ability to automatically run in-memory calculations in parallel. These benefits can provide an important competitive advantage to users within the financial services industry and in other application domains.
Unlike traditional map/reduce implementations which process large, file-based data sets, Grid Computing Edition stores data sets in ScaleOut StateServer’s scalable, distributed cache as a large set of cached objects. Grid Computing Edition automatically maps the user’s “map” operation across both multiple cores within each server and multiple servers within the grid, and the map operation then executes in parallel on a queried set of cached objects. GCE’s runtime system maximizes parallelism and minimizes data motion within the grid, In addition, GCE performs high-speed, parallel execution of the user’s “reduce” operation across all grid nodes. The result is very high performance for the complete map/reduce operation.
Importantly, the user does not need to write special code to harness the power of the grid’s infrastructure. Grid Computing Edition’s APIs enable map/reduce calculations to be written as in-memory methods that avoid explicit cache accesses, and it provides automatic parallel speedup. In addition, the need for traditional HPC message-passing within the user’s application is eliminated.
“Grid Computing Edition represents a breakthrough in automatically scaling the performance of single-server, memory-based, map/reduce computations. It also opens up an important and exciting new role for distributed caching in existing grid computing environments,” said Dr. William L. Bain, founder and CEO of ScaleOut Software.” The fact that this solution is extremely simple for users and compatible with a wide range of grid computing environments will make it applicable to a wide range of data-parallel applications.”
Having been popularized by large search engine vendors, the map/reduce design pattern is well understood and is a perfect fit for calculations which need to analyze many data sets in parallel and then combine the results into a single report. For example, Monte Carlo simulations used in portfolio analyses are often written with a map/reduce structure to provide huge performance benefits. This programming structure is also fits well within a Web service that receives requests for analysis and returns results back to the requesting clients.
About ScaleOut StateServer®, ScaleOut GeoServer®, and ScaleOut StateServer Grid Computing Edition™
ScaleOut StateServer provides distributed, in-memory caching for server farms and compute grids to boost application performance and offload database servers. Its patent-pending technology for scaling performance and replicating data enables it to deliver scalable access to cached data while maintaining high availability in case of server failures. Tests have shown that ScaleOut StateServer’s performance quickly outpaces database servers as the load on a server farm grows. By using ScaleOut StateServer’s distributed, in-memory cache for application data storage, developers can maintain fast response times while their server farm grows to handle increasing workloads.
ScaleOut StateServer runs on Linux, Solaris, and Windows servers and communicates over an existing LAN or grid-interconnect. Once installed, the product provides powerful APIs that enable Java, .Net, and C/C++ applications to cache data in a single, uniformly accessible, distributed cache. It can also be used to replace ASP.NET’s built-in session-state storage and enable ASP.NET applications to seamlessly and transparently store and retrieve session-state. ScaleOut StateServer implements comprehensive distributed caching semantics, including distributed locking, object timeouts, and automatic memory reclamation. In addition, it offers a unique asynchronous event handling mechanism which automatically distributes the event handling load across the server farm and reliably delivers events even if a server outage occurs.
The ScaleOut GeoServer Option uses data replication to extend distributed caching across multiple, geographically distributed data centers so that they can share fast-changing workloads and be fully protected against site-wide failures. GeoServer’s capabilities help IT managers meet the stringent performance and uptime needs of high-end Web sites and other mission-critical applications.
ScaleOut StateServer Grid Computing Edition uses breakthrough technology to provide automatic parallel speedup for data-parallel, “map/reduce” computations operating on data sets held in ScaleOut StateServer’s distributed cache. GCE’s use of straightforward, in-memory applications opens up the power of grid computing to a wide range of developers.
About ScaleOut Software, Inc.
ScaleOut Software develops software products that provide scalable, highly available storage for workload data in server farms. 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 at daveb@www.scaleoutsoftware.com or visit www.scaleoutsoftware.com. ScaleOut Software, Inc. 10900 NE 8th Street, Suite 900, Bellevue, WA 98004, T: 503-643-3422.
#####