By using website you agree to our use of cookies as described in our cookie policy. Learn More

TDWI Upside - Where Data Means Business

Why Enterprises are Liberating Their Mainframe Data

Liberating mainframe data and enabling newer and more agile analytics activities is becoming increasingly common and is no longer an unusual or experimental practice.

Mainframe organizations face many challenges, but the current pandemic laid bare the vulnerability of mainframe environments when a business suddenly needs to support and enhance their flexibility and adaptability. There was no inherent problem with the mainframe, but its traditional pace of operations and often-rigid operational schedule hinders its potential effectiveness. In other words, what was missing was deliverability.

For Further Reading:

The Surprisingly Strong Case for Mainframe-based Analytics

Bringing Data Back Home to Big Iron

Spark Comes to IBM's System z -- But Who's Buying?

Many organizations rely on the solid reliability of the mainframe and on a rhythm of regular reporting and analysis exercises to inform decision makers. However, the pandemic demonstrated to many organizations that the traditional pace of mainframe wasn't fast enough to help address numerous ad hoc challenges.

For example, a major North American transportation company had a regular cadence of moving a subset of DB2 data off the mainframe once a day. It was a well-honed system that was fully adequate in normal times. However, when operations needed to be refocused and reconfigured from moment-to-moment to get key shipments where and when they were needed, the traditional approach was insufficient. Suddenly, the company needed to understand where all vehicles were located in order to reroute and reprioritize as global supply chains slowed to a near standstill. Hourly data updates became a necessity.

Fortunately, the company had adopted mainframe-compatible software that didn't burden the CPU with either data movement tasks or ad hoc analyses. The company was able to assign the data movement tasks to zSystem Integrated Information Processor (zIIP) engines -- an often underutilized mainframe feature. With this hardware capability, the software routed the DB2 table data quickly and efficiently to target systems, enabling their non-MF data analysts to update data lakes with current DB2 data outside of the MF platform, making changes and generating new reports without needing to change MF processes. This low-cost capability allowed the organization to be unexpectedly adaptable. Instead of faltering under the crush of pandemic events, it actually adapted and improved performance.

The use of zIIP engines to liberate mainframe data and enable newer and more agile analytics activities is becoming increasingly common and is no longer an unusual or experimental practice, though in comparison with established legacy methods it can still seem novel.

Aside from supporting greater frequency of large data queries, it can also assist with the perennial problem of slow sequential access to data on physical tape or in virtual tape systems. Because accessing this data has presented such a burden to the mainframe central processor, the data is often underutilized and infrequently utilized. Data movement capabilities using software that runs on zIIP engines means this treasure trove can be accessed and moved (using lightweight extract, load, and transform (ELT) methods) so cloud or on-premises analytics can take advantage of information siloed on the mainframe. The same method for data movement can also be used to supplement or replace traditional slow and resource-intensive data backup and recovery.

Most mainframe organizations could benefit from better leveraging zIIP engines. Well-written, efficient, software-based solutions that can move data over TCP/IP to both on-premises and public cloud is a reality. Likewise, on-premises hybrid cloud solutions that support object storage connected directly to zSystems over TCP/IP allow more work in parallel and can enhance on-premises security.

These cost-effective technologies are available today -- while the price of inaction only grows over time.

About the Author

Gil Peleg, founder and CEO of Model9, has more than two decades of hands-on experience in mainframe system programming and data management, as well as a deep understanding of methods of operation, components, and diagnostic tools. He is a co-author of eight IBM Redbooks on z/OS implementation. He holds a B.S.c in computer science and mathematics.

TDWI Membership

Accelerate Your Projects,
and Your Career

TDWI Members have access to exclusive research reports, publications, communities and training.

Individual, Student, and Team memberships available.