As 70% of Fortune 500 companies continue to rely on the mainframe, it’s clear the world’s most reliable computing platform isn’t going anywhere anytime soon. Still, for the mainframe to continue to remain the most cost-effective source of computing power at enterprise scale, the software running its programs must continue to receive regular updates.
One of the most recent and remarkable modernizations is IBM’s release of COBOL V6—the latest in a long line of updates to a language that grew out of a Department of Defense effort and first saw use in 1959.
COBOL V6 has some significant improvements over its predecessors, including the ability to use modern mainframe hardware found in the z14 and z15 that COBOL versions all the way up through 4.2, released in 2009, could not capitalize on. For enterprises needing to leverage these latesthardware processing improvements, it’s more of a self-imposed mandate than an option. Thanks to 64-bit addressable space, COBOL V6 can also compile large programs that COBOL V5 couldn’t hope to compile even at OPT(0). For all its improvements over V4, many organizations have skipped V5 altogether.
With a new version of COBOL and the latest mainframe hardware, enterprises can enjoy reduced CPU workloads and a significant cost-savings over more outdated options. However, upgrading to COBOL 6 can be a particularly disruptive proposition for many other organizations.
Despite Its Advantages, the Obstacles to a V6 Update
COBOL V6 offers some major advantages to the right customers but getting an application to run on this updated language obviously means recompiling. Depending on your existing mainframe hardware, this step might be cumbersome (and costly) enough to negate any benefits associated with the updated version of COBOL.
COBOL V6 also comes with some drawbacks, including compile times that can be incredibly drawn out compared to the same procedure on version 4.2. Why the increase? Even at the minimum optimization level of OPT(0), COBOL V6 performs a more complex series of computations than those performed by version V4.2. For a program that sees fewer compiles and far more executions, it’s a smart tradeoff—an increased upfront investment that will yield dividends down the road. On the other hand, for an application that’s compiled frequently but rarely runs in production, the optimization might increase costs to a point that negates the return.
Alleviating Upgrade Pains
If your organization is on the fence about a COBOL V6 upgrade, you should strongly consider leveraging a Compiling-as-a-Service (CaaS) methodology. Compiling in the cloud allows you to drastically reduce costs by circumventing IBM’s process-based licensing method and allows you to pick the LPAR to recompile on. By letting Cloud Compiling take care of your compiling needs for you on our cloud servers, we can take that burden off your plate altogether. We’ll also keep track of all the load libs and copy books from your past compiles, and the additional capacity in our data centers can help you scale at a moment’s notice.