PLATFORMS: Windows NT with a Microsoft SQL Server database.

BACKGROUND: Sun Chemical is the world's largest supplier of printing inks, organic pigments and graphic arts materials, with 1997 worldwide sales expected to exceed $3 billion. Sun Chemical is a subsidiary of Dainippon Ink and Chemicals, Incorporated of Tokyo, Japan.

PROBLEM SOLVED: The data marts have been utilized for two purposes at Sun Chemical. Sales related data marts support on-line analytical processing (OLAP) for sales and marketing decision support and contain both low-level and summarized data. Other data marts are used as operational data stores (ODS) to support our industrial-strength, complex-page, financial and manufacturing MIS reporting. We looked to data marts to address performance problems for our OLTP systems that also were utilized for query reporting and decision support.

PRODUCT FUNCTIONALITY: The Sagent Data Mart Solution's data movement technology handles all the data extraction, transformation and population of our data marts. Sagent's Data Flow Plans consist of data sources, data transforms and targets. Design work is done within the client piece of the product--the Design Studio--but the actual running of Plans occurs on the server--the Sagent Data Mart Server--and is completely independent of the client. All plan and administrative data is stored in the integrated Sagent Repository. The Sagent Data Mart Solution's unique iconic programming and Data Flow Plans helped us deploy our first data mart in just four weeks--and we did not pick an easy one to begin with. "Programming" consists of dragging building block icons from the toolbox, connecting them and specifying parameters on a block-by-block basis. (It doesn't feel like "programming" at all!) Although the catalog of transforms delivered with the product is quite robust, the ability to easily extend the catalog with reusable, custom Visual Basic and C++ transforms was a key factor in making our first project such a quick success.

STRENGTHS: Results from multiple queries, even from multiple databases or flat files, can be combined in a single plan and fed to the target data mart. Complex joins that go beyond the capabilities of simple SQL can easily be accomplished within a Sagent Plan. The multi-purpose Expression Calculator transform makes it easy to convert data types, perform arithmetic, manipulate strings, and incorporate IF-THEN-ELSE logic in plans. Some transforms, such as the Key Generation for populating fact tables, or the automatic generation of the Time Dimension are designed specifically to support the population of star schema structured data marts and are significant time savers. Using the Grid transform to peek at the data stream from any point in the Data Flow is very helpful during Plan development. The actual data is always on the screen for instant verification.

WEAKNESSES: Sagent's biggest weakness is the lack of a robust facility to capture, store and report on meta data.

SELECTION CRITERIA: After reviewing the available tools for data mart population, we selected Sagent because it is very aggressively priced and because we can accomplish 99 percent of our goals without the need for procedural code.

VENDOR SUPPORT: The quality of support from Sagent rivals the best in the industry.

DOCUMENTATION: The bad news is that the user manuals lack detail descriptions of the transforms and also need to be brought up-to-date for the latest release. The good news is that the product is generally so intuitive that you don't need the documentation at all. It would also be helpful to have a meta data document on the Sagent Repository itself. The table and field names do not lend themselves to easy interpretation.

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