Sybase, MySQL, IBM boost data management

Real-time notifications, clustering touted

Database vendors Sybase, MySQL, and IBM are lacing products with features that improve access to information and offer clustering and fault tolerance.

Sybase and MySQL are improving data management wares. Sybase is releasing real-time data management software and a new version of its relational database, and MySQL is adding clustering to its open source database.

IBM confirmed that the upcoming Stinger version of its DB2 database will be available in a beta release in June. It will feature Partition Advisor, which automatically partitions databases, and Information Integrator, software that offers information integration and analysis.

Sybase RTDS (Real Time Data Services) is intended to provide insight into changing business events. It provides a data management offering in which time-critical management data is pushed from heterogeneous enterprise databases to messaging infrastructures, Sybase said.

According to Sybase officials, RTDS has been equipped to provide proactive notifications of database events -- for example, when changes happen in the database.

Sybase’s replication technology can read database logs without being intrusive, said Haridas Nair, director of marketing for data management at Sybase. “You can capture events from these database systems without being intrusive. And we are able to deliver these messages, these events that we capture through a messaging system like JMS [Java Message Service] or MQSeries,” he said.

RTDS functions with databases from Sybase, IBM, Microsoft, and Oracle. Included within RTDS are several Sybase technologies, including ASE (Adaptive Server Enterprise) 12.5.2, which is a new version of the company’s database; Replication Server, for data replication; RepConnector, a tool to bridge Replication Server to messaging architectures; and a JMS bus.

ASE 12.5.2 features Smart SQL cache for better application performance, improved security through Kerberos, Pluggable Authentication Module, and Active Directory support (for 64-bit Intel platforms for increased Linux scalability). Additionally, the database has been made a consumer of Web services.

Looking to boost its presence in enterprise applications, MySQL last week made a clustered version of its open source database available in a preview version, with the production version set for release in the third quarter of 2004.

MySQL Cluster combines the MySQL database with a clustering architecture for 99.9 percent availability for mission-critical applications, according to the company. The database features a distributed in-memory clustering architecture to boost availability and throughput, with response times of between 5 and 10 milliseconds and a throughput of 100,000 replicated transactions per second on a four-node cluster with two CPUs per node.

MySQL is offering MySQL Cluster under its dual-license business model, in which it is provided at no cost under the free software/open source GNU GPL (General Public License) for open source projects and under a commercial license for software vendors and commercial customers.

MySQL Cluster runs on operating systems such as Linux, Microsoft Windows, Sun Solaris, and Mac OS X. Hardware platforms covered include 32- and 64-bit Intel systems, PowerPC, and Sparc.