INDUSTRY

Benchmark demonstrates linear scalability of Tangosol Coherence Data Grid parallel processing: Intel, IBM, and Tangosol have completed grid-based benchmarks that demonstrate the linear scalability of parallel processing of data using Tangosol Coherence Data Grid running on a computing grid of 100 IBM servers based on the Dual-Core Intel Xeon processor 5100 series. A U.S.-based investment bank, with more than $600 billion in assets under management, initiated this benchmark test to establish a performance and scalability baseline for Coherence. This Tangosol customer, a public company listed on the New York and Pacific Stock Exchanges, has been using Coherence since mid-2005. Coherence is the distributed in-memory data grid backbone for several trading and mission-critical applications at the company. The requirements for the benchmark identified by the bank included: - reliability and high availability; - linear scalability (scale-out); - very high performance (scale to the bandwidth of the network); - ease of use and extensibility. The benchmark results showed that Coherence enables linear scaling of parallel processing across the grid as the grid (and the data set) increases in size and is limited only by aggregate CPU cycles across the grid. In the test, Coherence was able to linearly scale from 2 million aggregations with two servers to more than 60 million aggregations across the 96 servers. This 30 times increase in processing throughput was achieved with only one tenth of a second increase in processing time, or 1.2 seconds compared to 1.1 seconds. Additionally, the tests demonstrated that the data grid storage capacity increases linearly as additional resources are added to the grid and is limited only by the amount of RAM available to the data grid. "These tests indicate that companies using Coherence Data Grid can achieve unprecedented performance on industry-standard hardware, for operational and cost advantages," explained Cameron Purdy, CEO of Tangosol. "Because Coherence is so finely optimized, it fully utilizes all resources by moving the data closer to the applications - in memory - where it can be available instantaneously as demand grows." "This benchmark of Tangosol Coherence Data Grid successfully demonstrates predictable scalability," noted Elliot Garbus, general manager of Intel's Developer Relations Division. "As the industry moves from dual-core to quad- core processors like the new Quad-Core Intel Xeon processor 5300 series for data grid servers and the new Intel Core 2 Extreme quad-core processors for data grid clients, enterprise customers will be able to achieve even greater performance and energy efficiency." "This internal benchmark proves that the improvements to today's computing hardware combined with capabilities of products like Tangosol Coherence both validate and greatly enhance the power of grid computing," said Tamara Crawford, Program Director for IBM's Grid Ecosystem Strategy. "IBM is pleased to partner with companies, like Tangosol, to demonstrate the benefits of grid technology, to extend the scope and capabilities of IBM Grid offerings and to speed enterprise adoption of grid computing." Tangosol also performed parallel aggregation and processing tests across the grid, which demonstrated a linear scale model allowing for such processing to take less time as the grid grows, or to process more data as the grid grows.