eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
TimesTen Inc. this week will unveil an upgrade to its in-memory database that boasts enhancements in performance, scalability and ease of integration with larger relational DBMSes.
Version 5.0 of TimesTen Real-Time Event Processing System, available now, adds support for Intel Corp. Itanium 2-based servers running Red Hat Inc.s Red Hat Linux, officials said.
For the upgrade, the Mountain View, Calif., company tweaked concurrency—the ability for multiple users to access and share data simultaneously—to improve performance, officials said.
Instead of requiring data managers to use locks to prevent multiple users from changing data at the same time, TimesTen 5.0 uses a multiple-version model where users can query and read data at the same time that the system processes updates.
TimesTen 5.0 doubles the previous versions capacity for CPU use to a maximum of 90,000 transactions per second, officials said.
Version 5.0 also offers enhancements to Oracle Connect, a built-in set of features that allow the product to be used as a real-time cache for data stored permanently in an Oracle Corp. database.
|
The update is designed to automatically push data into its permanent home in such an Oracle database—a feature that comes in handy when the product is used in call centers, officials said. For example, real-time call center work, such as routing customer calls, is done in TimesTen, while changes to the customers record are automatically stored in a permanent, written-to-disk relational database.
One TimesTen customer found the databases improved scalability a compelling enhancement.
Dynamicsoft Inc. embeds TimesTen in its software to handle real-time call processing. Steve Roskowski, vice president of marketing for the Parsippany, N.J., company, said being able to push through more data will increase transaction throughput, a big enticement for Dynamicsofts customers.
“The key driver for us is it reduces deployment cost for customers,” Roskowski said. “Instead of having to go to additional machines in a cluster for performance level, [customers] can go with a reduced amount of machines.”