There's no doubt that iSeries users believe the midrange server is being pushed out of the data center. Sometimes...
they're able to persuade their CIO or CEO to keep it, but more often shops are replacing iSeries servers with Windows-based machines.
Why is that? Some say it's because executives believe the iSeries is out of step with the mainstream, that Windows is the standard. Others say it starts with the high schools and colleges, which ignore the iSeries and expose students to only Windows technologies. And some say IBM isn't doing enough to educate people about the qualities of the platform.
Here's what a few Search400.com members had to say about the future of the iSeries and why it's losing ground in data centers.
Price/performance of other platforms better
What I have seen is the price/performance of other platforms is better than that for the iSeries, particularly the iSeries compared with the Windows platform. Even though in recent years that gap has been narrowing. Another factor that's affecting the iSeries platform is that the user community is more familiar with the Windows platform. That familiarly is being driven by colleges and high schools. For most users, their initial exposure to computers has been with PCs.
American Health Care Software
South Burlington, Vt.
Can't ignore the stability of the iSeries
Recently I sat with the senior management staff defending the existence of our AS/400. The staff felt the Microsoft assimilation was inevitable and that we may want to consider getting on board now. This idea born with the CEO drew the yes-men at the table in quite quickly. The CEO -- a self-proclaimed recovering Luddite -- ignored them. He wanted to know why we should stay with our legacy hardware (AS/400 & RS/6000). I had then, and have now, a one-word answer to all of this: stability. For those of you who don't know what that means, ask your IT staff and IT managers how often they need to reboot their NT, Windows 2000 or .NET servers? Our policy was once per week on the older servers to avoid the infamous BLUE SCREEN OF DEATH.
What about the RS/6000 running AIX or the AS/400 running OS/400 V5R1? Reboot once a month? No. The AS/400 was IPL'd (rebooted) only for testing or applying PTFs. The operating system is rock solid. IBM continues to embrace the user market with support for Linux partitions on the AS/400. If you need a cheap server solution, use Linux. If you need an easy server solution, use Windows 2000 or even .Net. If you need a rock solid server solution, Big Blue was then and is still now the answer I give when a company tells me they need 100% up time. I bet my job on it.
Technology services director
Active Computer Consulting Inc.
IBM needs to adapt the iSeries to needs of the market
The death of the IBM System/38 was predicted on the day of its debut back in 1979. This has continued all during the 15 years of the AS/400's existence.
If the market finally abandons the iSeries, it will not be because of the hardware architecture. It will be because IBM may have failed to adapt the system to the present needs of the market. That is, to be able to run different flavors of operating systems on top of the extraordinary base operating system, hidden in the iSeries.
I know IBM has been working on that for many years and that the job is nearly done for its own operating systems (AIX, z/OS) and for Linux. And in the past eight years, several rumors predicted the same for Windows (not an IXS).
If this can be done, then the future for the iSeries (or whatever it will be called) is secure. If not, then the system will slowly fade away as new generations of young IT people take over. They are growing up with Windows, are trained in schools that use Windows and know nothing else. Values such as stability, security, scalability and even compatibility mean less to this new generation because they are used to failures and loss of data. They want to use what they are used to and what everybody else is using. And above all, they want GUI, Web and all the goodies of today's computing capabilities.
IBM knows that. They have to act and act fast.
AS Consulting BVBA
Windows, Unix poor performance ignored
Unfortunately, I do think the iSeries is being forced out of the data center. I have no idea why, although I have some theories.
My employer places a huge amount of faith on Windows and Unix platforms. This despite the fact that hardly a day goes by when one or the other does not breakdown and cause a huge commotion, the platforms' lack of processing power, the inability of databases to be used in multiple, parallel ways, and the sheer inconvenience of using the system to begin with. None of those things seem to make any impression.
It takes us three to seven times longer to do a particular task using a Unix or Windows platform than it would take to do the same task on an AS/400 -- and one person can do it. Part of that is due to the fact that we have outsourced software services and server management.
The iSeries is no longer a strategic platform here and will be phased out within the next couple of years to be replaced with the new Unix servers that Sun has pretty much given away a few weeks ago.
I have my own theories about why the iSeries is being pushed out. One is that schools provide more impetus to Windows, Oracle, Unix, etc., pretty much ignoring the iSeries (indeed, any other "older" platform). As a result, people coming out of college/school have no idea what an iSeries is or they have been taught to fear or deride such platforms. While they may know Java and HTML, they know no COBOL. Once these people reach positions where they can make decisions, naturally they lean toward platforms they are familiar with.
The other reason companies turn to other platforms is that new graduates are in relatively high supply -- and probably cost less. It makes sound financial sense to buy a cheap Windows server and pay a new grad $30,000 rather than find an iSeries programmer who would command a bigger paycheck.
CIBC Wood Gundy
Brokerage Technology Operations
Execs say Windows servers are the de-facto server standard
I have to agree with the analysts that voiced the opinion that Windows will be the dominant server platform in the next five years. In our shop, our primary corporate back-office application is running on four iSeries servers. However, in the last year we have replaced an iSeries server running Domino/Lotus Notes with three Windows 2000 servers running Domino/Lotus Notes. We are also in the process of replacing three Novell print/file servers with three Window 2000 servers. Finally, we are also in the process of installing three new Windows 2000 servers running Web applications.
What is the reason for these replacements you ask? It is our leader's opinion that the iSeries and Novell servers were out of step with the "mainstream" IT shops. It is our leader's opinion that Windows servers are the de-facto server standard for the IT industry. It is also our leader's opinion that it is easier to recruit people with a Windows skill set than for either the skill set of Novell or iSeries.
Decision makers ignorant of Windows' flaws, iSeries superiority
I think iSeries is the best computer that money can buy. But IBM is the first company to ignore it. I have sold and programmed and installed iSeries systems from the beginning. I saw it always getting better, but often I also saw it being pushed out because the decision makers know Microsoft, they have experience in doing restarts, and they don't know anything about iSeries. They are not aware of security and stability problems in peer networks (a lot of dwarfs will not be able to replace a giant). That's all result of half-hearted marketing by IBM. iSeries always has been the stepdaughter in the zSeries and storage-minded company. And so I fear iSeries will go down in the next few years.
becom Informationssysteme GmbH
IBM needs to push open technologies even more
The contenders for the data center floor space: iSeries, mainframe, Windows and Unix. Those can be divided into two camps: EBCDIC and ASCII.
It seems that if anything holds the iSeries back it is the basic data "incompatibility." Windows and Unix are clearly in the same camp with the PC or personal workstations. Mainframe and midrange support staff don't mind the conversion, but as volumes increase on the Unix and Windows side, the ease of data transformation in our homogenous environments will be the key, both for data transformation and application porting.
For VCRs, Beta was a good format, but the average consumer preferred the ease and availability of VHS. I loved Apple's PCs, but the company I worked for and the majority of my friends all had IBM-compatible PCs. When it came time to buy a new system, I bought what they had so we would be compatible.
Porting has issues other than data. Between operating systems certain assumptions have been implemented. To many programmers the differences appear minimal, but to advanced programmers the differences between dead-lock detection and lock timeouts can be a major problem. Java does help on both the data side and the programming environment, but adoption has been slow.
I do not see this pattern changing unless IBM and iSeries staffs can hide the EBCDIC nature of the operating system and push open technologies even further (Java, XML, Messaging) to compete head-to-head with the "Wintel platform" dollar-to-dollar and application-to-application.
Perhaps IBM can change the computing models by creating a truly "Magic Box" that combines the strengths of OS/400 with the openness of the pSeries (and Linux). A "Magic Box" that supports portable code, eases porting of existing applications and data, plus delivers more computing power for the same or less money as Wintel.
Otherwise, the iSeries loyalists will use these great iSeries systems just like the Apple Macintosh users did and do in this PC-dominated market. Even as the rest of the market continues to reboot their data centers.
Ensure applications remain available for the iSeries
When I worked for the local school district a number of years ago, we made hardware decisions based on the availability of appropriate software. I think that is a key issue with the iSeries and any other hardware. As long as there are applications that meet the needs of customers on one set of hardware that is better than on another, businesses will make the decision to use the best software. In our sector, the iSeries is the most appropriate hardware for now. IBM would do well to work with development companies to assure that good applications remain available for the iSeries platform.
Manager of user operations
New Mexico Educational Assistance Foundation
Senior managers see the light
Since our software provider offers the same applications on all platforms, my boss considered going to Windows and getting rid of the old iSeries. I wrote a report about the advantages and disadvantages of doing so, and it turns out there are a lot more disadvantages. There is no doubt in my mind that the iSeries is more secure and reliable than anything else. As a result, we upgraded to a new iSeries and everybody is very happy with the speed of that new box. We also have a Windows network that we use to connect to the iSeries, e-mail and Internet access. And that's all it will be used for. Running key applications on Windows is suicidal.
Walker Foods Inc.
Users don't like green screen applications
In our company I think the ISeries will eventually be pushed out of the company completely. We seem to be putting application responsibility upon the user departments, and they do not understand the AS/400 and definitely do not like the green screen applications that our packages run most efficiently on. Users want the point-and-click capability.
I am an old-timer in the midrange arena, but I must admit that most point-and-click applications are a neat way to work -- and the new, younger employees understand it better than the green screen. I also think that we are losing a lot of in-house technical knowledge of what goes on behind the scene because of the point-and-click functions available.
"Experts" lack expertise
I still don't understand how the so-called "experts" continue to push Windows like it's the operating system of the future -- even though they know there are far better operating systems than those coming out of Microsoft. They know that NT -- or should I say "DOS" -- continues to be enhanced with elementary services that anyone with a computer science degree would have learned and that must be present from day one and not in future releases. These common services have been, as your article presented, part of REAL operating systems for a LONG, LONG, LONG time, yet they get no attention.
Just because bicycles are used by the majority of people on earth, it does not mean that they will become the standard method of transportation. Unless, of course, Microsoft is the company building the bicycles. In that case, they and the "experts" will tell us that engines will be present within the next five years, and they will be asking companies to wait for that release.
This industry is loaded with so called "engineers" (i.e. a six-month class on how to point/click and take options from good looking menus). These "industry experts" have only known bicycles and cannot understand that there are already bikes, cars and planes. It is beyond their six months' training given to them by Microsoft. After all, isn't Microsoft in the cutting edge of technology? Why even bother to finish high school? "I got my six-month certificate, and I'm an engineer!"
The fact is that until we get this industry back to real experts, i.e. computer scientists who have experience with a number of operating systems and are aware of the many options for operating systems and so on, this industry will continue to evolve backwards.
In the transportation industry there was a process of evolution from the wheel to carriages to bikes to cars and so on up to the shuttle technology. With computers we skipped a lot of the evolution process and got right to computers with operating systems designed to manage multiple tasks without interfering with each other. And what do we do? The follow up: Design an operating system that can barely handle two or three tasks at the time before giving you the blue screen of death. From planes to carriages!
The perfect slogan for Microsoft should be: Buy our product today, and you will see how the NEXT release will be so much better.
You have to give it to their marketing team: THEY CAN SELL YOU ANYTHING! (Of course after buying it from some one else.)
IBM to blame if iSeries becomes extinct
Let's face it. If the iSeries goes extinct there is nobody to blame but IBM. The sales of an iSeries machine is nothing more than a short-term profit to them (IBM) because the iSeries is so stable they can't sell their Global Services to go along with it. I wouldn't be surprised at all if IBM is going to let it die for that reason.
It's like having a light bulb that burns forever. The only people that benefit long term from it are the ones who bought it.
MidTech Consulting Services
Small shops can't afford to rewrite for another platform
We began running our proprietary application (COBOL by the way, no RPG) on an AS/400 model 50S in 1997, our first midrange machine. We upgraded it to a two-way model 720 a couple of years later, and we're still using it. We're very happy with it for two main reasons:
2. Built-in database
While I'm as puzzled as most about IBM's iSeries marketing strategy, I wonder if small companies such as ours may form the basis of a foothold of sorts for IBM in that our application would take years (read "high development costs") to rewrite for another platform/language suite. Therefore, we have no compelling reason to move off the AS/400. It would cost more to rewrite/rehost the application than to keep the AS/400 or its iSeries successor. We plan to upgrade the hardware in 2005, probably to an iSeries model 810. It seems likely that there are many other small companies who are in the same situation -- very content with their AS/400-iSeries and want to stay that way.
John D. Davis, III
Manager, Computing Services
Texas Insurance Checking Office Inc.
People stick with what they know
No matter how much or how effective IBM's advertising is, IS/IT people simply do not want to change platforms no matter how dependable or easy to use or more beneficial a new platform might be. I've been a paid IS pro since 1985, and it's pretty obvious that whatever people have learned -- be it in high school, college or their first IT position -- that platform/language is what they usually stick with and defend to the death. It's the old Ford vs. Chevy thing. My first car was a Chevy, so "they're the best and Ford sucks."
Large insurance company