The Relative Merits of IBM and Teradata Data Warehouses

By Rich Hughes,

A recent report concludes IBM® PureData™ System for Analytics (PDA) has distinct advantages over a comparable Teradata Data Warehouse appliance.  The International Technology Group (ITG) determined the three year cost of ownership of Teradata’s 2750 as 150% more expensive than IBM’s PDA N200x.  ITG surveyed 17 Teradata and 21 IBM customers, and matched their production data warehouses by workload, system size, and industry.  As an example, both Teradata and IBM customers in this study were categorized into Telecom, Digital Media, Financial Services, and Retail industry sectors.  ITG then further built their survey profiles based on data volumes, number of users, and workload similarities within each industry category—thus promoting reasonable comparisons based on real world observations. Starting with bottom line cost-of-ownership, the initial system expenditures for each customer industry profile were in the same ball park.  That means that IBM’s PDA lower cost advantages were found in deployment, support, and personnel costs over the systems’ life time.    For the application deployment area, IBM’s PDA delivered the lion’s share of their applications in 20 days or less.  As the chart below shows and by stark contrast, none of the Teradata customers deployed as quickly, with the Teradata average deployment being more than three months. 1

Rich Picture1

Getting an analytics application deployed months earlier speaks to time-to-value, and attaining more business objectives. The simpler to use IBM® PureData™ System for Analytics appliance proved to be less expensive to operate because of fewer full time employees managing the system.  The Teradata Data Warehouse appliance requires more administration training to understand how to achieve performance goals.  For the ITG study, the 17 Teradata organizations average 1.7 full time employees to oversee the Teradata data base, as compared to less than 0.5 full time employees to tend to IBM’s PDA, on average.  The primary reason for this significant gap is that IBM’s PDA end users are able to directly access their analytic appliance as a self service data warehouse. One of the ITG study participants using both Teradata and IBM’s PDA summed up the differences this way: “…we don’t have to build indexes… users write directly to the system, they don’t need to go through a DBA…we work with complete data sets instead of having everything aggregated and summarized first…we don’t have to use data models. In comparison with Teradata systems, performance-tuning overhead was said to be virtually non-existent.” 1 ITG determined that IBM customers realized additional benefits versus Teradata because of the ease in which end users deploy applications, creating “… the potential for closer business alignment than conventional data warehouse techniques.”1  In a world where vendors make significant claims of performance and total cost, it is hard to argue with real customer experience. To learn more about ITG’s findings, the full version of ITG’s report can be found at 1.

About Rich Hughes,

Rich Hughes has worked in a variety of Information Technology, Data Warehousing, and Big Data jobs– and has been with IBM since 2004.  Hughes earned a Bachelor’s degree from Kansas University, and a Master’s degree in Computer Science from Kansas State University.  Writing about the original Dream Team, Hughes authored a book on the 1936 US Olympic basketball team, a squad composed of oil refinery laborers and film industry stage hands.  Hughes is an IBM Marketing Program Manager for Data Warehousing, and you can follow him on Twitter at @rhughes134.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s