Data Directions: A Decade of the AS/400

Over the past 10 years, as I have written for various magazines, given talks at Common or some other user group about the AS/400 and its technology, one topic is always on everyone's mind. When will the AS/400 die?

It has no native GUI, it is not current with technology (whatever that means) and heck, it’s run by IBM, they can't do anything right. So why are we still concerning ourselves with this thing?

If you read most publications outside of our little world, the AS/400 is dead, taken over by Unix. No wait, that was a few years ago. Today, I mean NT.

If you think there is something wrong here. There really isn't. It's just a natural course of events in the IBM midrange world. It has always been tough to get much respect from those outside of the IBM midrange arena. But the trouble is that most of those in today’s AS/400 market, love it.

So here we are again spelling the demise of the AS/400 as NT and Microsoft get the press and, more importantly, the unit sales. However, those of us who have been around a while have heard it all before. Other systems – like Wintel -- are just now talking about their 64-bit architectures that will be out in the coming years, the AS/400 has had it for years. In fact, IBM has released four versions of the 64-bit operating system.

As far as technology, no one can claim IBM is at the bleeding edge of state of the art, but then again, my clients can't afford bleeding of any kind. Nevertheless, while Unix, then NT, claimed that it was the only way to go, the AS/400 quietly delivered most of what the others just talked about.

Now we have databases from a number of vendors touting their capability to handle large databases with millions of records. Of course they are still trying to figure out how to back them up, but they are going in the right direction. New versions of data warehouse tools are coming out on all platforms with their biggest problem -- performance in large-scale environments and connections to the legacy data that actually does all the work. Then SQL Server announces it will soon have record level locking on the database. Wow -- the AS/400 has quietly had it since the CPF days of the System/38.

The negative press never ceases, while big technology watchdogs claim that the AS/400 is dead and IBM can not keep up with what is required. Here we are with the most successful midrange platform in history, and yet the pundits claim that the product is a dud. Are we missing something?

Actually, I think the pundits are missing something. The companies who use the AS/400, while not technology backward, are not concerned with technology as a goal. Their goal is to run a profitable business with the best technology that would help them at a cost that made sense. And guess what? It still is. The AS/400 is successful because it was designed to help customers do exactly that.

That does not mean that we have no leading-edge technology on the AS/400, just that it is not the driving force. I have had many discussions with IBM and others at user groups about what we need on the AS/400. It is critical that IBM keeps up with the technology that companies are adopting. In fact if they don’t, the AS/400 will die. But IBM and Rochester in particular, has shown a resiliency when burned by something new by quickly recovering and bettering what others have done. Hmmm…sounds like a certain Bill I know.

Ten years after the AS/400 was first produced it is still going strong and a $4 billion dollar annual business is not going to go a way anytime soon. More importantly, it is a business that gives us the best operating system on the market with the best hardware platform. Oh sure, I wish it was a little cheaper. I wish I had a native GUI. I wish I did not have to integrate NT with it. I wish I could figure out IBM's marketing and support plans and I wish, oh how I wish, that the media outside of our little world would take a look at see that the force to reckon with the mighty Microsoft is already here.

The AS/400 may not be able to play Starcraft, but it does a pretty mean order entry system.


John Bussert is president of Swift Technologies (Marengo, Ill.), a company specializing in AS/400 and Windows NT software.