Sunday 12 February 2012

TimesTen Upgrade Delivers Improved Performance

By Ron Williams


Oracle recently released TimesTen In-Memory database 11d release 2. Also, it is expected that Exalytics In-Memory machine will be launched before long. The price for the product is available on Oracle's website. According to Oracle, the TimesTen up grade gives you overall performance benefits, and scalability that will enable it to support a larger number of users at a time, take a much bigger volume of requests and handle considerably more sessions. As it's one of the two main components needed to run Oracle Exalytics (the second being Oracle Essbase) TimesTen will not just be limited to transaction processing but will likewise be able to support analytic applications as a result of its brand-new adaptive caching abilities.

Highlights of TimesTen

This brand new caching element, when coupled with Exalytics and the Business Intelligence Foundation Suite, allows TimesTen to assist with analytic apps. When it is working with OBIEE, the TimesTen adaptive cache can look at the workloads that'll be created by numerous queries and dashboards. Adequate information about these tools is crucial, and that's why professionals should be thinking about enrolling in obiee classes and staying up to date with the changes. Anytime the cache can find a response time blockage in an OBIEE application or in apps which are meant to run OBIEE as middleware, the information is going to be sent to the memory of the Exalytics system. The Exalytics system has 1TB of RAM and it's powered by four E7-4800 Intel Xeon processors which have 10 cores.

To many of you, 1 TB may not be adequate RAM however TimesTen also will give you compression in columnar form. This means that 1TB is basically equal to 5TB of addressable memory. As per Oracle, the data will be scanned in the main memory in a matter of 5 seconds. If you want to raise the system's total capacity, then you must increase the number of Exalytics boxes. TimesTen's caching capabilities are also adaptive in nature because they go on modifying and improving themselves based on the changes in dashboarding and query demands. In addition, they depend more on the in-memory speed, materialized views, and other techniques that are more dependent upon administration work than on database.

TimesTen and Hana

TimesTen has primarily been a part of the transaction-processing arena, since it has been used within that segment since the latter part of the 1990s. It was used by telecommunications providers, on-line retail, as well as in financial buying and selling for low latency apps as an embedded database. SAP's Hana experienced a very different starting point. Hana was initially put to use in analytics as an accelerator to run the SAP BW (Business Warehouse). Rather recently, Hana was used as a method to run analytic applications as well as to replace the BW database.

Nevertheless, now both of these in-memory technologies of SAP and Oracle are about to collide. The key reason Hana poses a difficulty to Oracle is because SAP is marketing it as a product that will prove to be an in-memory replacement of BW running databases. Because of the large client base which SAP has, and with 60% of the instances of BW using Oracle database, just about any modifications to Hana could affect Oracle adversely.




About the Author:



No comments:

Post a Comment