Maybe the “i” Is for “Indecisive”

Article ID: 21021

Most IT organizations like to consolidate. They want fewer computers, fewer types of computers, fewer hardware vendors, fewer software vendors, and so forth. This quest for less makes good sense because the more you can standardize on a limited number of vendors and products, the easier life gets. Systems and software management becomes easier, vendor management becomes easier, even help desk calls become easier — all because there are fewer options and therefore less complexity.

Interestingly, the same benefits of consolidation apply to server vendors. If these vendors can find ways to consolidate their offerings, they also enjoy benefits, although slightly different ones. If, for example, a server vendor could produce one system used for multiple purposes, instead of multiple systems each used for one purpose, that vendor could simplify its supply chain, its manufacturing operations, its support infrastructure, and more.

Thus it should come as no surprise to you that this is exactly what IBM is doing . . . sort of.

Don't be fooled by the recent press release: The convergence of System i and System p has been going on for quite some time. Remember when the AS/400 came out in 1988 and used a proprietary CMOS chip? And remember when, in the mid-1990s, the AS/400 moved to the PowerPC processor along with the RS/6000? Even better, remember how many people thought that this move was the end of the AS/400 — that the AS/400 would become just an option on the RS/6000?

Clearly, that didn't happen. Time passed. RS/6000 became pSeries and then System p. AS/400 became iSeries and then System i. Both systems moved from the PowerPC processor to POWER4 to POWER5 and now POWER6. And at virtually every one of these processor changes, people asked, "Will the AS/400-iSeries-System i be replaced by the RS/6000-pSeries-System p?" And the answer, thus far, has been no.

What about this latest announcement, however? Is IBM taking a step much closer to the final convergence of these two platforms? The answer to that is yes and no. But let's put it into the context of this column — "hot and not hot."

Hot: IBM Reinvests in System i

If nothing else, IBM's reshuffling of System i responsibilities into "enterprise" and "SMB" divisions shows that the System i is still important to IBM. And it should be. System i continues to be a cash cow for IBM. But over the last few years, IBM has had a hard time positioning the System i in the SMB market, where x86 servers are the dominant life-form, and also in the enterprise market, where they compete against the System p and other midrange offerings. Frankly, it seemed to me that IBM was much more interested in selling System x and System p solutions to these markets than System i solutions.

This reorganization lets IBM better focus on these two important market segments. In the SMB market, the System i is often an ideal solution, but IBM has been doing a poor job of making the case for this, allowing Windows-based x86 servers to grow like weeds in the SMB community (not that I have anything against Windows-based x86 servers). On the enterprise front, i5/OS has a lot to offer, but once again IBM has had a difficult time positioning it against AIX. By combining forces here, IBM should (at least theoretically) be able to position the two offerings in a more balanced way.

I certainly can't see how things could get worse in either market segment.

Not Hot: IBM Can't Close the Deal

Although the realignment of System i resources is, for the most part, good news, it's a little frustrating that IBM can't just suck it up and collapse these two systems together. We've all known for over a decade that IBM will eventually make one line of hardware platforms for both AIX and i5/OS. We could all happily get on with our lives if IBM would just do it.

But IBM can't seem to get to that point. It is one baby step after another, one reorganization after another, one processor changeover after another — each change bringing us inexorably closer to hardware convergence. Come on, IBM. Just do it!

Astro's Non-Pick of the Litter

I wish I were writing about the iPhone this week. I've been watching the TV commercials and thinking about how easy it would be for me to nose my way from web pages to location maps to phone dialing. Think about it — this phone is the ideal phone for a dog to use. But did my master get one? Oh, no, he did not (even though he had several chances to do so). He grumbled to me about no support for 3G services, complained to me about being unable to replace the battery, and told me he was going to wait for the 2.0 version. How cruel my master is!

Sean Chandler is a computer network consultant with over 30 years of field experience. Astro, a border collie with more than 40 dog years of data processing experience, provides technical support to his master, Sean. You can reach them both at schandler@SystemiNetwork.com.

ProVIP Sponsors

ProVIP Sponsors