Podcast: If Integrated Correctly, Legacy Mainframes are Powerful Business Intelligence Tools

February 16, 2009

Ensuring timely delivery of new applications in a cost effective manner is critical in today’s challenging economic climate. Organizations must continue to look at all dimensions of new solutions, namely the impact on the people involved, the process for integrating older legacy assets, and how to implement the infrastructure to support the solution to meet the required service levels. The simplest way of ensuring success for the business unit is to ensure that the process is simple and quick, enabling frequent interaction and review between the business unit and IT.

Listen to the Podcast:

Integration legacy mainframe assets into a modern business intelligence environment introduces unique challenges in that the architectures used to build mainframe systems decades ago and very dissimilar to modern data and application infrastructures. To ensure speed and quality, we need to present legacy assets to developers as though they are familiar – we need to ensure that we introduce industry standards to systems that we developed prior to the creation of those standards. This will enable developers to present that legacy data quickly and accurately to the business unit.

Mainframe systems are characterized by their use in high volume, reliable, and predicable transactional environments. Organizations need to be able to develop new solutions non-invasively, namely without requiring changes to existing mainframe systems. Introducing changes triggers the entire software development lifecycle for mainframe applications, testing, quality assurance, performance and scalability impacts etc. These associated delays and costs must be avoided to meet the goal of faster less expensive new solutions.

zIIP (System z Integrated Information Processor) specialty engines have been introduced by IBM as a mechanism to provide a more cost effective platform for computing. zIIP specialty engines differ from the standard General Purpose Processors (GPPs) in that:

They cost about 10% of a GPP.
They run at maximum electronic speed rather than governed or pre-set speeds.
Workload executing on zIIPs runs software license fee free.
Only specific types of mainframe workload can be dispatched to the zIIP.

BI and SOA-based applications provide the cornerstone to success for most organizations – the simple reuse of existing intellectual property enabling improved decision making. The heavy lifting required by middleware to support these critical environments demands that we address the needs of all participants – people, software and hardware.

When exploring new solutions on the mainframe it is critical to ask the question, “Does this have the ability to take advantage of the high speed, low cost zIIP specialty engine?” DataDirect Shadow is a great example of being able to transfer 80-90% of its workload to a zIIP specialty engine, providing high performance low cost mainframe integration. The combination of presenting mainframe assets, data and application assets, through industry standards on a low-cost, high-performance platform assists organizations in meeting their goals: