In the era when 75 percent of the systems are developed in Java, PHP or. NET, hundreds of companies still base their core infrastructure on IBM i (or as the old litter we call it… AS/400).
These companies are mostly financial institutions and manufacturing that evolved from the ’70s, some migrating from the old IBM S/34, S/36, S/38 and then commiting to the great promise of late ’80s IBM AS/400 (Application System 400). What was so attractive for companies to enter in the 400 world? It was very simple: IBM promised that everything that worked on S/36 and S/38, will still work on AS/400, but more efficiently. This means that moving to this platform would incur minimal cost from the point of view of development for companies that decide to migrate, and at the same time, remain under the «Standard» of IBM. At the same time, Unix was still an operating system more oriented to the academic world than business, and Intel-based servers were really in its infancy, so the «Mid Range» market (which did not justify the cost of a mainframe and at the same time did not have enough with what low-end computers offered ) was seduced very convincingly to become an AS/400 user.
This was intended to be a success story, had it not been for the late ’90s began a technological revolution that IBM had not planned: Microsoft began to settle as a possible competitor with the release of Windows NT Server to companies.
Let us agree that IBM was so blinded to their business that missed opportunities which, in the opinion of several experts, could have stopped the progress of their competitors. An IBM AS/400, while not as expensive as a mainframe, was more expensive (both hardware and software) than their counterparts at Intel. Students and independent engineers could not afford a piece of hardware like this, while IBM did not generate sufficient agreements with universities so that their own academies could provide these computers to «hungry» technology researchers. But hey… they could pay Intel machines.
Miss this opportunity, was a heavy blow to IBM and its Platform. Scholars became professionals and their companies employed technologies different to AS/400. The increased demand for these technologies introduced a lot of capital to the companies that produced them, and became direct competitors: growth of intercommunication between the platforms, development of open standards, Internet … all these points were developed without having AS/400 into account. By the time IBM realized this, his popularity was already very low. Worse still, they didn’t really made an effort to integrate these technologies until mid 2000.
For more IBM tried to force changes in its platform (enhancements to RPG programming language, integrated environments through PASE, including name changes, to move from AS/400 to iSeries, then System i5, and now IBM i), the popular view remained tied to «AS/400».
Today we often listen terms like SOA, XML, Agile Methodologies, Web 2.0 … but those terms are mostly specialists away from the IBM i. While systems require faster changes every time, integration with other platforms and services, and a brand new Web front end, the AS/400 was framed in the collective unconscious as a «Green Screen» platform. Increasingly, managers of companies with IBM i think «Why can my competitor have a sales system so dynamic, web screens accessible from their cell phones, and I only have to offer to my business a Green Screen?». Others may think «Why does my computer department take so long to build this or that report, and when they made it and I ask for a change they say it would take at least a week to do it?». Never heard such comments? Well, the answer (no matter how crude it seems) can be summarized in these points:
• While the «open» systems (based on Java,. NET, PHP, etc …) were released since a relatively short time (no more than 10 or 15 years), it is common for systems that are running today in an AS/400 that they have great pieces of code written over 30 years ago… when the requirements and vision of what comprised a computer system were completely different from today.
• IBM, with all the possibilities of backward compatibility, and its separation from the academic world, indirectly make most specialists in the platform reluctant to changes and new technologies («If it works fairly well … for why touch it? «).
• There is also a matter of age: According surveys of AS/400 professionals (http://www.teknodatips.com.ar/encuestas/18-as400/156-encuestaedadas400.html), 50% of the current iSeries professionals will be retired within 10 years, so they are people who are more concerned with how to «round» their last years of profession, than to learn new technologies.
SH%$@T! This means that the picture is so unfavorable that I should leave the platform? NO! These are just some of the facts. Let’s see the other side of the coin: What is positive with AS/400?
• While most of the developments on open platforms are trying to standardize their databases, AS/400 platform was the first operating system integrated into a comprehensive relational database (DB2/400).
• While most of the Open Systems integration is being performed in Middlewares (Message Queues, WebServices, etc.). AS/400 was the first, also from its operating system, to manage the concept of Data Queues, DDM, and ICF, which is the basis of what later evolved into standards such as JMS, Web Services, and RPC.
• While the current system infrastructure is standardizing concepts such as Virtualization, Virtual Machines, and Containers, AS/400 was the first to integrate LPARs, the Machine Interface and Subsystems, also within its the Operating System.
• The cake cherry: Most of the Core Software of companies that currently have IBM i, is in the IBM i.
So if AS/400 integrated so many years the origins of these concepts, which are the basis of modern software architectures … obviously it can’t be all lost.
The advice I most often give to my clients is make the most with what they now have: If human resources are AS/400 aged over 40 years, and they know the Core system by several years of ongoing maintenance, it is really unproductive to try to «modernize» them… but they have a huge «Know How» of how business operates, and which things are necessary to transform their systems. And above all, they know the user and their needs very well. Do not try to transform them, take them as functional analysts because they are the best resources you have to get this task done.
And what to do with the new coming requirements, and with the managers requests? The secret is integration. The «i» of the new names of the platform (iSeries, System i5, IBM i), stands for integration. The golden opportunity offered by the IBM i today that can be fully exploited is to integrate the technology thank «always worked well», with agile methodologies that get faster results and implementation.
I’ll never say this is easy … but I can assure you from personal experience that this is the most effective way. The purpose of this blog is to give an overview of the steps to enter the to the XXI century maelstrom, without getting rid of our «AS/400».