Ask TPM: Enticing Users to Upgrade Their i5/OS Hardware

Published: November 5, 2007

by Timothy Prickett Morgan

Our experts over at Four Hundred Guru get asked a lot of questions and they provide a lot of good answers. But some of your questions are of a strategic or tactical nature, having to do with broader issues in the information technology space or with the i5/OS and OS/400 platform. So if you have a question you want to get my take on--or you want to use me to get a straight answer out of the people in IBM who know what is going on--then send me an email and I will see what I can do. Here's a questions I got last week concerning how IBM might get i5/OS and OS/400 shops to upgrade to newer System i iron.

Hey, TPM:

I was reading your story from last week's issue on IBM's upgrade planning tool (see IBM's Math on User-Priced System i Versus Vintage Machines) and one sentence jumped out at me: ". . . This box works, and it is not a problem for me, like so many others. Why should I upgrade it?"

Here's the question. Does IBM consider that if the functionality of i5/OS is enhanced, users will purchase new systems to a) have access to the new features and b) have enough CPU cycles to run the enhancements?

Let me give you an example. What if the system journalled everything: recreated programs, source members, all database files, system values, exit points, config settings, and so forth so every program, application, source member, and library could be dialed back to the state it was in at an arbitrary time in the past? How much excess capacity and increased functionality would that give users?

If you use your imagination, you can come up with a number of CPU-guzzling i5/OS enhancements that would entice customers to purchase all sorts of IBM iron.

--Steve

Hi Steve,

Yeah, no kidding!

Like many of us out there in the i5/OS and OS/400 community, I have banged my neurons together to figure out such a means to absorb the increasing computing capacity IBM delivers in the AS/400, iSeries, and System i product lines, which is especially important when you realize that Moore's Law is delivering increasing computing performance at a rate that far exceeds the normal rate of growth for transaction processing capacity at many i5/OS and OS/400 shops. Each processor generation represents an even bigger jump, which translates into an ever-increasing amount of time between upgrades for most customers.

IBM doesn't just have a marketing problem with the System i, but a workload problem. Which is why the company has been focusing on logical partitions and server consolidation for so many years, with a certain amount--but not nearly enough--success.

Five years ago, in November 2002, I put together my multi-part iDeal iSeries stories to show IBM how to create, market, and position a different kind of iSeries product line based on the then-new Power4 processors, which did not even get to the iSeries until May 2003 even though they debuted in the pSeries in October 2001. (Grrrrr. . . . )

At the time, when I was looking at IBM's quandary of ever-increasing CPWs of power in a machine, stagnating or moderately growing workloads among customers, and a desire to keep charging a lot for an iSeries server, I suggested taking a chunk of the CPW capacity in each OS/400 server--like half--and making a shadow logical partition that ran high availability clustering software behind the scenes. (See The iDeal iSeries, Part 4 for more on that.) The basic idea I put forth was to make HA a part of the system, like the DB2/400 database management system was, but unlike that example, allowing third parties to supply the HA clustering software. IBM could reduce the price on core 5250 and server capacity, charge a nominal fee for this shadow capacity, and help drive HA usage up dramatically in the installed base. This is a volume HA play instead of a value play.

Did IBM listen? Hell no. But, to its credit, it began to listen to some of the criticisms we were all making about the high prices of the iSeries line and the way the box was packaged. Way back then, I said to shift to a mix of user-based and CPU-based pricing on software, and to get the initial configured, usable system with 30 users activated on the box down to $18,000. This year, only five years after I asked for the iDeal iSeries line, IBM delivered machines more or less like this as the user-priced System i5 515 and 525. IBM has solved the user-pricing issue, but not the capacity issue as far as I am concerned, and given that a lot of the OS/400 base seems to want to create applications using tools other than WebSphere, that excess capacity inherent in the Power5+ machines is not being used effectively. If IBM gave away an extra core of processing, but it drove lots of software and services sales, it would be well worth the trade. But try telling the bean counters at IBM's Systems and Technology Group or within the Power Systems division or the Business Systems division that. (Remember, there is no System i division any more.)

Other innovative IT companies understand this concept you are talking about now, Steve, and I was talking about five years ago and even earlier. Here's one relevant and important example. Apple Computer just started shipping its "Leopard" Mac OS X v10.5 operating system on October 26, which comes in desktop and server editions. Mac OS X is based on FreeBSD Unix, with the Apple graphical user interface slapped on top and an emulation environment underneath it from Transitive that allows applications compiled for PowerPC variants of Apple machines to run on the newer Intel X64-based machines. With Leopard, Apple has added a feature called Time Machine, which is an automatic backup utility that stores the entire state of your machine to an external storage device for future recovery. Considering how cranky PCs are and servers can be, Apple is probably going to get a lot of software upgrades just because of this Time Machine feature. Apple gets it--and it gets the money, too.

My basic concept, since I have been writing about the AS/400, has been to convince IBM to get its machines competitive by dropping prices or adding features that will be broadly adopted by forward-thinking companies--the kinds that deploy OS/400 and i5/OS--and that are not integrated features on Unix, Windows, and now Linux platforms. If you start with a price target, you can hit that. You can get people to spend that money. But the way they feel about spending that money can be radically different. My wife was thrilled to buy me an iPod Nano this week. The world's best-integrated server platform should be exemplary, just like an iPod and just like an AS/400 used to be.

I could not agree with you more, Steve. If the rest of you reading have any concrete ideas on how to rejigger the System i, don't be shy, send them to me by email--hit that Contact button above--and I will compile and share them with all of you and IBM, too.

--TPM

RELATED STORIES

Ask TPM: Two System i Questions, and Two Responses

Ask TPM: What About System/3X Compilers in i5/OS V6R1?

Ask TPM: The Economics of Open Source Software

Ask TPM: System i5--Good Investment or Not?

                     Post this story to del.icio.us
               Post this story to Digg
    Post this story to Slashdot