In advance of Sun Microsystems Inc.s JavaOne conference for developers, Sleepycat Software Inc. will announce on Tuesday the general availability of Berkeley DB Java Edition Version 2.0—although the company said the new version has already seen strong adoption by commercial and open-source customers.
Berkeley DB Java Edition stores data in the applications native format. It features full ACID (atomicity, consistency, isolation and durability) transactions and recovery, record-level locking, zero administration, and a simple, programmatic API.
Berkeley DB JE 2.0 picks up support for JCA (J2EE Connector Architecture), JTA (Java Transaction API) and JMX (Java Management Extensions).
"Given that were talking about a transactions database, JTA is the crunch standard to support," said Gregory Burd, product manager.
"With [JTA], we can participate in [Java] group transactions, … Thats critical for being able to do things transactionally for business systems. Without JTA, you really cant participate in a multistep transactional process in a [Java] transaction server," he said.
The other APIs make it easier to work within an applications server, Burd said.
Need for speed is also a major focus of Berkeley DB JE 2.0. Rex Wang, vice president of marketing, said that the hallmark of the product is that its built for high-performance, pure Java systems.
Storage efficiency has been improved so that the amount of redundant data in the log has been reduced. This has resulted in a performance gain of some 30 to 70 percent when the application needs data thats not in cache and has to retrieve it from disk, he said.
Sleepycat has also improved concurrency: Formerly, you couldnt deploy a pure Java solution that dealt with a very large database, Burd said. "You needed a high degree of concurrency," he said. "We believe with 2.0 you can store enormous amounts of data thats accessed concurrently."
As an example of 2.0s ability to cache large data stores, Burd pointed to users such as Archive.org, an Internet library that archives vast numbers of URLs for the purpose of offering permanent access for researchers, historians and scholars to historical collections that exist in digital format.
Archive.org uses Berkeley DB JE to crawl the Internet and cache the data it finds—a data store that currently numbers more than 50 billion URLs.
Support for 64-bit platforms goes right along with the high-performance theme, Burd said. "Thats in line with the general theme of people using the Java edition for high-performance systems, looking for speed, for J2EE support, and looking for support for 64-bit platforms."
Berkeley DB JE supports all versions of Unix, Linux and Windows as well as most other major operating systems. Its used by open-source projects including JBossCache, Apache Geronimo, JavaSpaces and Spring.
Berkeley DB JE 2.0 is available now under a dual license: A no-cost open-source license permits redistribution if the application using Berkeley DB Java Edition is open source. A commercial license is available for redistribution of proprietary applications. Berkeley DB Java Edition can be downloaded at Sleepycats site.