Mainframe > Tips & Techniques > Systems Management

A System Programmer's Guide to Data Warehousing

A mainframe systems programmer must address high-level issues when installing a data warehouse

My transformation into becoming a data warehousing systems programmer happened quite suddenly with a job transfer into the data warehouse SWAT team within IBM Software Group. Truthfully, I didn’t know what to expect, but it turned out to be interesting and challenging work. My skill set was not lost, but rather enhanced through observation and hands-on work with a data warehouse (DW). The “lessons learned” from this transformation are the basis of this article.

Many high-level issues must be addressed when installing DW and business intelligence (BI) software on the IBM System z* platform, from the perspective of the mainframe systems programmer. From this standpoint, three major questions must be answered:

• What is a DW, using a system programmer’s terminology?
• What do I need to do as a system programmer to put in a DW?
• How will my job as a system programmer change?

The Terminology

If you’ve been in IT for any length of time, you start to see the same things reinvent themselves, and in most cases they work much better than their ancestral counterparts. It’s no different with data warehousing. Early attempts at data warehousing were extracts–where an end user could download files to his or her PC and use a spreadsheet to crunch data into various reports and graphs. You might recall, in those days, how database administrators and systems programmers would get requests to massage and manipulate some of the data in these extracts. Today, using DW/BI vernacular, this process is known as extract, transform and load (ETL).

For the past several years the systems-programmer function has been evolving into a more full-service role.

Tweet