Continue in 2 seconds

How to Save on Data Warehouse Resources

Published
  • December 01 1998, 1:00am EST

The best way to predict the future is to make it up. And, that is essentially what data warehouses are doing today. These critical decision support tools help companies predict the future by providing knowledge of what has happened in the past. Your data warehouse is a competitive weapon and can be the foundation for your strategic direction. It is far too important not to wisely spend money on it; but make no mistake, there is a vast difference between added value and excess costs.

Some of the hidden and sometimes excessive costs in data warehousing include data maintenance, data reorganizations, data placement, system administration and query optimization costs. A common mistake made by many companies is to choose the wrong database or structure and then throw people at it to fix the problems ­ expanding in an ever-larger cost spiral.

Fortunately, I have been involved in doing training at hundreds of successful data warehouse sites, including Wal-Mart, Sears, Lucent and American Airlines. These successful data warehouse sites all have three things in common:

  • Teradata is the database of choice for their decision support data warehouse;
  • They keep the profits high and the expenses low;
  • They have thousands of users and only a few system administrators.

As leaders in their own industries, many of the companies listed started out with small databases that grew dramatically over time. Theses smaller databases were easily and quickly modeled, and each success stimulated growth. The now-larger databases using the tables that were modeled previously began to grow in database size and nearly double on a yearly basis.
Because of its unique strengths, Teradata is the database of choice in the majority of "big league" data warehouses worldwide. One of those strengths is that Teradata can save companies a significant amount of money. Here is how:

Power and Performance.
Teradata is the most powerful DS (decision support) database in the world. It is also the only database on the market that uses parallel processing to load, backup and process data. Parallelism is what enables Teradata to handle complex ad hoc queries, large amounts of data and hundreds of concurrent users in data warehouses that range from 10GB to the largest data warehouses in the world.

Scalability.
Teradata has linear scalability that enables it to perform at the same level at 100 gigabytes as it does at 25 terabytes. When the data warehouse is upgraded with additional hardware, the Teradata database automatically redistributes the data evenly so that legacy applications can immediately take advantage of the new parallelism and computing power.

Field-Tested Optimizer.
Teradata has the most widely field-tested optimizer in the DS arena. Queries rarely have to be optimized by a database administrator (DBA). Instead, the optimizer chooses the best join paths, with all queries accepted and optimized by it. When you combine power, speed and scalability of Teradata with a proven intelligent optimizer, you have the makings of a world-class data warehouse. This combination allows users to have detailed data in a data warehouse or data marts, or a combination of both. It also permits modeling of the data using third normal form. This will save your organization money, because the data is being modeled after the business.

Most databases don't have the power, scalability or optimization techniques to use third normal form, so they compromise the model to fit the technology. These databases are forced to denormalize the data or use star schema modeling. This will complicate the data warehouse effort and place the data warehouse in a vulnerable position, requiring perhaps 5-10 DBAs to spend time continually fixing it.

World- Class Utilities.
Teradata's data warehouse utilities also have automatic parallelism built in. These utilities convert inbound and outbound data, load directly from a mainframe or client with no separate file transfers required and have checkpoint/restart capability. In a corporate data warehouse, users need to be able to load data from multiple sites with speed and ease. And they must also be able to extract information from the data warehouse and have it sent anywhere in the enterprise. So, utilities such as these can save time, and their simplicity can provide significant savings in DBA costs.

Ease of Setup.
Teradata features one-step SQL statements for database, table, index and column creation, modification or deletion. For example, when Teradata was used to run the database, tables and indexes for the 17-query TPC-D benchmark, it only needed two pages of commands. Other databases required between 175-300 pages. Think how essential it is to be able to create and modify easily in a competitive and changing business environment.

Teradata can save companies money by performing and scaling efficiently with no complications or hidden costs. Its optimizer allows users to model and design the system using the best and most uncomplicated models. Queries never require tuning by a DBA. And, the field-tested Teradata optimizer provides the best optimization. Further, Teradata's load utilities and ease of setup also allow savings on maintenance, loading and prevent chaos. This allows your data warehouse team to focus on the business problems users have and their future applications that will enhance corporate decision making. A Teradata-based data warehouse enables your company to keep costs low and spirits high!

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access