Oracle Exalytics: Is It A Must-Have for BI?

Oracle's latest engineered system promises better business intelligence performance and ease of deployment, but the tradeoff is in openness to third-party tools.

Cindi Howson, Founder, BI Scorecard

October 11, 2012

4 Min Read
InformationWeek logo in a gray background | InformationWeek

The headlines from the main stage at last week's Oracle Open World event were all about cloud, but in the business intelligence space, the focus was on Exalytics, Oracle's in-memory engineered system for BI.

Released in February, Exalytics is Oracle's latest engineered system. It leverages the TimesTen in-memory database to support fast BI analyses. The system packs 1 terabyte of RAM, 40 processing cores, and has a list price of $170,000 (not including software).

Exalytics lets you cache both the metadata for Oracle BI Enterprise Edition (OBIEE) as well as the data from data warehouses and Essbase cubes into memory. Oracle customer Sodex reported at Open World that it's experiencing ten times faster performance on Essbase cubes running on Exalytics versus a conventional Essbase deployment. What's more, the results served up at high scale to 10,000 users.

Exalytics integrates with Exadata, Oracle's database appliance, via a superfast Infiniband connection to allow loads from the data warehouse into the in-memory BI system. The Exalytics Summary Advisor gathers usage statistics on queries to recommend which data should be loaded into memory versus left on disk.

[ Want more on Oracle's engineered system answers? Read Oracle Makes Case For Exalytics, Data Discovery. ]

Oracle touts Exadata as the ideal solution for high-performance data warehousing, so why would anyone need Exalytics? Oracle customer Thomson Reuters reported at Open World that Exalytics has enabled the company to serve up an interactive dashboard analyzing a 10-billion-row data set--something that would not be possible without in-memory performance. Exalytics ensures performance at the BI application level by caching OBIEE metadata, whereas Exadata's performance addresses only the underlying data.

Exalytics also brings advantages to OBIEE software deployments. Swedish shoe retailer NilsonGroup cited easier maintenance of OBIEE on Exalytic's optimized hardware, without the need to patch software as they previously had to on commodity hardware. With Exalytics, Oracle handles the updates at the system level behind the scenes.

Oracle's approach does beg the question: should customers have to buy an engineered system to ensure a smooth deployment? One integrator I've heard from on this topic said software quality lapses in OBIEE 11g delayed project timelines by months, and he described the security integration with Oracle's WebLogic application server as "a train wreck." As OBIEE (which is based largely on Siebel Analytics, acquired in 2005) has become more integrated with the rest of the Oracle product stack, Oracle has become more dependent on its own middleware, taking a less agnostic approach than other BI vendors.

Another contrast is the comparison between Exalytics and SAP's Hana. The latter offers third-party support for multiple BI vendors and multiple hardware vendors. An SAP and Oracle partner who has tested both Hana and the combination of Exadata plus Exalytics described Hana as "revolutionary and 1,000 times better" while calling the Oracle alternative "evolutionary and 10 times better."

OBIEE 11.1.1.6, released in February, supports greater dashboard interactivity and improved data visualization over the previous version of OBIEE to take advantage Exalytics performance. At first glance, the finished charts look as powerful as those in competitive products, such as Tableau Software and TIBCO Spotfire. However, the OBIEE design interface is less intuitive than those mature products and there's a whole lot more complexity involved in developing dashboards in OBIEE than in the competitive products.

In addition to running OBIEE and Essbase, Exalytics also runs Endeca Information Discovery software. Oracle acquired Endeca in late 2011 both for its e-commerce technology and its ability to search unstructured content, such as warranty claims and Twitter streams. The Endeca MDEX engine uses a combination of in-memory and columnar data storage. Endeca can be deployed on commodity hardware, but a system integrator speaking at Open World cited easier deployment on Exalytics.

Endeca has a very nice faceted search interface that combines the search simplicity of Google with the guided navigation of e-commerce sites. It was encouraging to see this niche product touted on the Open World main stage by Oracle CEO Larry Ellison, but it was frustrating that the product was not available in the BI demo pods. I got a tweet after leaving Open World that it was being showcased in a different building, not near the rest of the BI capabilities. Too bad.

The bottom line, if you are using OBIEE and hoping for high user counts or big data scale, Oracle has designed Exalytics to be part of that deployment. In my view it's still too early in Exalytics' product life to say how much customers are sacrificing openness to third-party products for the promise of performance and ease of deployment.

About the Author

Cindi Howson

Founder, BI Scorecard

Cindi Howson is the founder of BI Scorecard, a resource for in-depth BI product reviews based on exclusive hands-on testing. She has been advising clients on BI tool strategies and selections for more than 20 years. She is the author of Successful Business Intelligence: Unlock the Value of BI and Big Data and SAP Business Objects BI 4.0: The Complete Reference. She is a faculty member of The Data Warehousing Institute (TDWI) and a contributing expert to InformationWeek. Before founding BI Scorecard, she was a manager at Deloitte & Touche and a BI standards leader for a Fortune 500 company. She has been quoted in The Wall Street Journal, the Irish Times, Forbes, and Business Week. She has an MBA from Rice University.

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights