The Esper development team has issued the first production release of its open-source Esper event-stream processing and event-correlation engine. It will compete with commercial complex-event processing (CEP) offerings, including StreamBase and Coral8, that also analyze high-volume, low-latency data streams in real time.
Esper will ease development of applications that filter event streams and detect patterns, aggregate data over time- and length-windows, join streams and trigger actions based on the presence or absence of particular events, according to project lead Thomas Bernhardt. A software architect for a financial institution, Bernhardt says the open-source approach helps in "tailoring the event-processing language and other community-driven features" to both broad and specialized needs.
Most of Esper's rapidly growing community is still evaluating how best to exploit the software. Promising applications include real-time monitoring of proprietary options trading, network monitoring and use in RFID research. But Esper does not yet match the performance and capabilities of commercial event-processing engines.
The latest release of StreamBase handles more than 500,000 messages per second and supports up to 30 terabytes of data, according to Bill Hobbib, marketing vice president. Coral8 Version 4 handles comparable data rates with superior event-pattern matching, claims John Morrell, product marketing director. Just as for Esper, key StreamBase and Coral8 selling points include reduction in development time and cost achieved by moving away from custom code. StreamBase offers a graphical interface not available with Esper, while Coral8's preferred interface is its SQL-extending Continuous Computation Language. Both commercial products offer reliability features required by enterprise adopters but not yet available in Esper.
Users benefit from the spectrum of options. "With multiple players in the marketplace, the tide rises for everyone," Hobbib says. --Seth Grimes