By James Serton

When I wrote my previous article (see Downfall of the Mainframe – Still Alive and Computing,) my intent was to establish the foundation and groundwork for ensuring that the knowledge gets preserved on how to maintain and enhance the mainframe system, so that a new generation of mainframe developers and engineers can come forward to take over the tasks outlined. Now that we have established this foundation and ultimately proved that most industry pundits were wrong, the next question is, what do we do? Does the phrase “Gentlemen, we can rebuild him” mean anything?

In this circumstance, we can draw a unique parallel. In the “Six Million Dollar Man,” a series that aired on ABC in the 1970’s, (originally adapted from the best-selling novel, Cyborg,) ace test pilot Steve Austin’s test plane crashed, leaving him nearly dead. As a result, the government decided to rebuild him, augmenting him with cybernetic parts that gave him superhuman strength and speed. In return, Austin becomes a secret operative, dedicating his existence to fighting injustice where it is found.

Coming back to 2007, the phrase “Gentleman, we can rebuild him” has a unique meaning when talking about mainframes and coincidentally has been used more than once when discussing various topics about bringing older technology into the 21st century, especially now that we have transferred and established the knowledge needed to ensure the mainframe system will continue unabated. The key now is to take that knowledge and combine it with current software technologies and concepts, ultimately integrating the mainframe into today’s technology environment. While the past investment in mainframe applications can still pay off for a long time, broader opportunities exist to train younger talent in the mainframe domain as established with the mainframe competency approach and take advantage of their creativity to develop innovative IT solutions, ultimately mixing old world ingenuity with new-age concepts and developing a stronger population of mainframe systems.

Ladies and Gentlemen, we can enhance the CICS application beyond the conventional red, white and blue 3270 display (how patriotic!). Imagine if you will, running a browser; something we all take for granted in this day and age and retrieving data from a CICS application, then in turn bringing that browser screen alive with a fireworks display of audios and visuals never imagined on the old conventional 3270 terminal.

Today, the text representation (commonly known as the 3270 dumb terminal) while still used throughout the IT industry is in fact obsolete. Today’s user demands ease of use and integration (i.e. a click on a dropdown menu versus memorizing hundreds of codes). Marrying old world ingenuity with new-age concepts is fundamentally simple. Choosing from the hundreds if not thousands of different technologies, methodologies and solutions to accomplish the task is hard. For sake of argument however, the specific technologies and methodologies up to the task, which will be addressed in this article are 3270 emulation and Service Oriented Architecture (SOA.)

When one researches emulation as a technological alternative, the actual definition is clouded. However, in defining it for use within a mainframe, Merriam Webster provides the best definition of the term ‘emulation’ or ‘emulator’ when creating a relationship between today’s technology and mainframe systems. An emulator is defined as a piece of hardware or software that permits programs written for one computer to be run on another computer. In the case of the mainframe, 3270 emulation software can effectively duplicate the functions of an IBM 3270 terminal on a PC (Windows, Mac, Linux, etc.) machine, allowing the developer to access TSO, CICS and other mainframe tools and use them in concert with today’s technology to develop a new generation of browser-based applications that can be utilized to retrieve data and present it in a more aesthetically pleasing interface, whether a Mac or PC, complete with the capability of generating visualizations such as charts, graphs and more – functions never thought about in the 70’s and 80’s.

One of the many products on the market that can emulate the 3270 terminal for development purposes is HATS by IBM, which stands for Host Access Transformation Services. In its simple form, HATS transforms the 3270 green screen into a browser window with the vibrant colors of today’s current display technology. Used in conjunction with such technologies as JAVA, a developer can create web applications by integrating mainframe data with data from other platforms. This marriage of technologies can tap into the raw power and bring out the true potential of the mainframe, ultimately bringing the mainframe alive with new capabilities going beyond the conventional mainframe system.

HATS and other like-minded products provide the answer to what tool can be used to ‘rebuild them.’ But before development can commence, how can we ensure that standards are ultimately adhered to and that a mainframe system can talk to a PC, a Palm Pilot or other device? The answer lies in a methodology that has ultimately been around throughout the inception of the mainframe, but is only now catching traction. Service Oriented Architecture (SOA) can provide the mainframe programmer with the opportunity to use his or her knowledge to develop or orchestrate business services, consisting of self-contained units of work compromising a recognizable business task. Since the mainframe developer is familiar with the necessary business functions, he or she would be responsible for the definition and functionality of these packages. These business services can then be packaged and published to the remainder of the non-mainframe development team in the form of APIs or XML objects which in turn can be used by the JAVA or .NET developer and integrated into other distributed systems applications.

One of the key benefits of this approach is the standardization of the data being utilized by all downstream applications. This concept as stated, revolving around the re-usability of code, is not new in the mainframe world. Mainframe software libraries are full of subroutines that are published for other developers to use. These are business packages developed by those that are most familiar with the particular functionality of the application being packaged. Why re-invent the wheel when all the application data and information is already present? Those who have come before us have already figured out how to manipulate dates or pars 20 digit account numbers along with a multitude of other factors. These functions do not need to be coded again and again. Ultimately, a well-defined SOA environment will help lay the groundwork for standardizing business services available from any particular mainframe environment. Mainframe developers that understand the appropriate applications from when they were developed and enhanced will have the upper hand to develop these business services.

So once again, we have established through the transfer of competency within the first article and the capabilities of enhancement that the mainframe is alive and computing. Utilizing these practices and techniques makes the mainframe even more valuable that it ever was. The abundance of data available from the mainframe will never cease and using the right mix of tools and techniques will help to bring new developers into the realm of understanding the business process that resides in the minds of the outgoing mainframe developers. This step (a big step) is one of (if not the most important aspect of) developing a new mainframe competency rising from the knowledge gathered by those that are leaving the industry.

James Serton is project manager for DATA, Inc. www.datainc.biz