6 December 2004 -- ObjectStore, a leader in products for real-time data services, and an operating company of Progress Software Corporation (Nasdaq: PRGS), today announced a major update to ObjectStore(R) Trading Accelerator(TM), which provides financial trading houses with real-time market data capture, query and management, integrated with historical market data. With its newly available Event Query Language (EQL), ObjectStore Trading Accelerator allows trading houses to apply sophisticated "what-if" scenarios and analytics on incoming data sources and execute such real-time trading strategies as volume weighted average pricing (VWAP) analysis and program trading.
ObjectStore Trading Accelerator features a "tick database" that can track market data from market feeds such as Reuters, Bloomberg and other financial information services. It captures this data at speeds up to 50,000 events a second, with sub-second latency. Financial traders can use this data in simple EQL queries to compare values with historical and descriptive reference data. For example, a trader may define a trading rule that generates an action when a series of values are attained or breached (e.g. generate a buy order when "a specified equity Price is greater than $75 and Volume is greater than 10,000 over the last ten minutes").
"ObjectStore Trading Accelerator delivers a trading architecture with the performance and scalability so crucial for trading houses seeking to move from T+1 to straight through processing (STP)," said Peter Sliwkowski, president, ObjectStore. "By delivering the ability to analyze and execute in real-time against existing and historical market data, ObjectStore Trading Accelerator ensures that financial traders optimize their trading processes for both speed and accuracy."
Facilitating Real-Time Trading and Market Data Management
ObjectStore Trading Accelerator has the ability to simultaneously collect, organize and query real-time data by leveraging three main components:
-- Query Server provides a native EQL interface for real-time analysis of event stream data. Trading applications written in C++ or Java(TM) can call the EQL interface, either directly or through JMS using SonicMQ(R), without performance impact on the event capture process.
-- Collector captures the event data supplied by the feeds in real time and inserts it into a persistent store for greater recoverability than traditional in-memory systems.
-- Archiver organizes and indexes the data for efficient use, and re-orders out-of-sequence data in the background for maximum performance.
ObjectStore Trading Accelerator is built on the ObjectStore Event Engine(TM), a real-time data management product for capturing streaming event data. The ObjectStore Event Engine provides event-driven applications with a reliable way to manage high-volume, high-velocity data and concurrently execute real-time queries on those event streams. The Cache-Forward(TM) Architecture (CFA) of the ObjectStore Event Engine provides the high-performance, in-memory data capture, data persistence and ACID properties needed to ensure the availability, reliability and recoverability of financial trading applications. This architecture also delivers the scalability needed to handle high-speed, high-velocity market data at in-memory speeds while maintaining a multi-terabyte data store of event data. ObjectStore Trading Accelerator can execute
analytics in real time against captured event stream data, such as real-time VWAP. In conjunction with ObjectStore's caching technology, it enables traders to leverage portfolio or other trading reference data, providing support for in-memory concurrent query access.
Availability and Pricing
ObjectStore Trading Accelerator can be deployed on a variety of platforms and is currently shipping. Pricing starts at $50,000 per CPU for development licenses.
