Dealing with the multitudes of legacy TPSs

Abstract

For the most part, our military's legacy Test Program Sets (TPSs) are written in some derivative of IEEE Std 416/716 ATLAS (called adapted subsets, e.g. CASS ATLAS). Some have been in use for 30 years. As the hosting ATE systems are wearing out, we are faced with upgrading the hardware to support the legacy systems as well as technology insertion via system upgrades or new systems/subsystems being designed. From a software perspective, there are many avenues to be taken to “modernize” the legacy TPSs for the future. : Stay with the legacy adapted ATLAS subset. : Re-host in an industry standard ATLAS system. : Translate into another standard such as ATML with a “Carrier” language. : Rewrite from scratch into “C/C++/C#”. : Re-implement in a bench tester “Graphic/drag and drop” type system. : Or some combination of all of the above. The goals are the same no matter which method is chosen. Create a suite of TPSs that will support the legacy “black boxes” and future “black boxes”, and that are maintainable for the next 30 years. This paper discusses the various approaches and some of the pitfalls we may encounter.