Business Intelligence Network business intelligence resources

Industry Research

« Data Management Buyers Want Performance and Maintainability First and Foremost | Main | Disecting the Magic Quadrant »

Selected Data Warehouse Vendor Ratings

As I have cautioned several times, Evalubase's research method comprises an ongoing study of the enterprise technology market. While our method is imminently more empirical (and far less magical) than "quadrant" type charts and doesn't include any vendor supplied data whatsoever, some of our data is quite preliminary. We caution vendors in particular about spouting any of these findings without checking with us first.

That disclaimer out of the way, here are current overall technology and vendor ratings for solutions/vendors predominant in the data warehouse market. Vendors offering business intelligence, data integration, data administration/modeling, data quality and metadata management solutions were selected for this comparison.

Overall technology ratings comprise user-supplied ratings of functionality, performance, scalability, maintainability, usability, security, portability, ease of integration, ease of implementation, satisfaction and value. Overall vendor ratings comprise user scores of the vendor's credibility, responsiveness, ingenuity, support, vitality, sales process, marketing, legal & accounting functions, licensing practices, and services & training capabilities.

At present, Actuate, Embarcadero, Ascential, Business Objects, Data Advantage Group, Firstlogic and SAS enjoy an early overall lead. Following them are Computer Associates, MelissaData, Siebel, DataFlux, i-Way, Information Builders, Informatica, Microstrategy and Cognos.

View Evalubase Chart

Many of the current scores are actually pretty close. We'll see what happens over time as you and your colleagues participate in the ongoing study. In time, we'll compare particular attributes of key competitors and study how (for example) their customer support ratings and ROI numbers evolve over time.

See you at the TDWI and Wilshire/DAMA events this month!

Just telling IT like it is,
Doug

Be sure to make your voice heard, get instant peer comparison scorecards, and qualify for complimentary access to all Evalubase research. Submit a confidential evaluation or two of your favorite (or least favorite) BI products today: www.evalubase.com

  Posted by doug laney on May 16, 2005 3:59 PM |

Comments

The following are some key factors about IBM newest version of DB2 (IBM UDB ESE 8.1) that provide a strong contrast with Teradata capabilities.

IBM ESE DBAs have lots of things to configure and maintain—and that task never ends. Teradata does those things automatically—no mistakes that can cause performance problems.

The IBM ESE Optimizer has a lot of parameters that need to be set. Parallelism is based on optimal configuration parameters. The Teradata Query Optimizer requires no configuration; queries are always fully parallel. The "ability" to "tune" queries is NOT a good thing--the system should do this automatically.

IBM ESE users typically do not access the warehouse directly. It can't handle it. This means the IBM topology duplicates data, creating redundancy and potentially unreliable data. Teradata users directly access the warehouse, concurrently, with complex ad hoc queries--no problem.

IBM ESE users typically need to have the DBA help set up new queries. That isn't needed with Teradata.

IBM claims 3NF but recommend Star Schema for ESE. Teradata recommends 3NF, but can do Star Schema.

IBM ESE needs a lot more disk space than Teradata for raw, unduplicated data, increasing expense, management work, and processing.

IBM ESE has much less built-in fault-tolerance and RAS than Teradata; and has single points of failure. Teradata fault tolerance is built in.

IBM ESE does not offer production-worthy load utilities. Teradata does. Data loading is a major concern of IT as it is an ongoing activity to build and maintain the data in the warehouse.

With IBM ESE creating a mixed workload environment- what we call Active Data Warehousing, is extremely difficult.In essence, IBM can not do short tactical queries while running complex ad-hoc queries while running loads at the same time.

With IBM ESE scalability of concurrent access is an issue because tuning is required. This is not an issue with Teradata.

Posted by: Richard Crounse | April 11, 2006 6:30 PM

Post a comment