It's Time to (Mostly) Retire Historical Mainframe Photos

325px-Candlestick_phoneChris Gaun at Gartner asks "Can Mainframes Be the Least Expensive Option?" (Well of course they can!) But I must take issue with his editor who decided to illustrate Mr. Gaun's blog post with a photo of an IBM 704 mainframe taken in 1957.

What's the freakin' point? Shouldn't illustrations have some reasonable relevance to the story? (And why stop there? How about a picture of Charles Babbage's difference engine? Or an abacus? Good grief!)

Here's a proposition. If every news editor starts illustrating all Apple iPhone, Samsung Galaxy, Nokia Lumia, and RIM BlackBerry stories with photos of analog cellular "bag" phones or rotary dial "candlestick" phones, and if every editor illustrates all stories about Apple iPad and Google Nexus tablets with photos of a 12th century monk's manuscript, I'll drop my objection to illustrating mainframe-related stories with photos taken years or decades before I was born. Do we have a deal?

Unless you're specifically writing about an IBM 704 and its use in 1957 at NASA, that's a really, really dumb photo to choose for your story about the popularity of 5.5 GHz IBM zEnterprise mainframes in 2013. Update: NASA didn't actually exist until 1958. However, NASA is the source of the photo Gartner used. The photo might make sense accompanying an article about NASA's computing history. That isn't Mr. Gaun's article.

Smartphone, circa 1928.

by Timothy Sipples January 29, 2013 in Analysts, Blogs, Media
Permalink | Comments (0) | TrackBack (0)

IBM Reports 4Q2012 Earnings: Full Speed Ahead

IBM reported its 4th quarter and full year 2012 financial results yesterday. Wall Street seemed impressed with the results, propelling IBM's stock price over 4% higher in after hours trading.

I like to take a look at the zEnterprise hardware results which IBM generally reports not because one quarter is particularly important, and not because IBM's hardware sales are all that indicative of the growth in the overall mainframe ecosystem. That said, they can give us some clues what's going on.

The clues are great. IBM's zEnterprise servers were the standout in the whole earnings report, with revenues (i.e. sales) up 56 percent in the 4th quarter (year to year). That's not necessarily a complete surprise since the 4th quarter was the first full quarter of availability of IBM's new zEnterprise EC12 model. However, Wall Street had some concerns that customers would hold back on their purchases due to economic uncertainties. That didn't happen with respect to IBM, it would appear. Also, IBM pointed out that zEnterprise revenues grew a whopping 68 percent in their growth markets (the countries outside the United States, Western Europe, and Japan). That portends well for zEnterprise growth, because those economies are growing faster. More zEnterprise customers in more countries growing faster is great for the entire mainframe ecosystem. (I would advise mainframe professionals to be aware of opportunities that might be outside their own borders.)

"MIPS," a measure of mainframe computing capacity, increased 66 percent. I've pointed out this simple math before, but here it is again: revenues were dramatically up, and capacity shipped increased even faster. That means IBM continues to offer lower mainframe capacity pricing to its customers. One of the common industry myths is that the mainframe market is a monopoly. Not in my view, and declining unit prices is consistent with a highly competitive server market. IBM is the most successful enterprise server maker, and there's no better enterprise server than zEnterprise (sporting those wonderful 5.5 GHz cores and extremely high qualities of service), but IBM keeps increasing its price competitiveness. That's exactly what I want to see at least as a long-term trend: IBM succeeding (to keep funding strong research and development efforts for even better products), and customers enjoying a healthy share of that success with both progressively higher business value and lower prices ("value for money").

IBM also said that they shipped the greatest amount of mainframe capacity they've ever shipped in history in the 4th quarter. Yes, some of that growth is attributable to "organic growth," which is the natural increase of transaction and batch volumes among existing applications. However, keep in mind that much of the developed world is in recession, and economic activity is not growing terribly fast there. So I think it's very fair to assume that much, maybe most, of this capacity growth is attributable to new applications, migrations of existing applications that were running elsewhere onto zEnterprise, and new customers. IBM didn't provide quite that level of detail, but I think that's a reasonable conclusion.

IBM doesn't break out its software and services businesses to identify which parts are attributable to zEnterprise. It's probably impossible to precisely calculate a "What if IBM didn't have its zEnterprise servers and somebody else did?" counterfactual. But most analysts conclude that IBM's zEnterprise business is essential to driving IBM's high profitability.

There have been some commentators that suggest that IBM has been de-emphasizing hardware, but I think that's simply wrong. IBM seems to continually reinvent itself, moving to where the ball will be before it gets there, not where it has been. I think hardware will always be a critical ingredient in that success, and more importantly so does IBM if you look at the company's research and development. There's an interesting comparison with Apple here. Apple got out of the X86 server business because, for Apple, that wasn't a business in which they could compete with high value, differentiated products. Apple also got out of printers and digital cameras several years ago, again for the same reason. But is Apple a hardware company? Or is Apple de-emphasizing hardware because Apple keeps distributing more music, movies, electronic books, "apps," and cloud services? Well, yes... and no. It's substantially the same with IBM, albeit in a completely different information technology market segment. (Apple and IBM literally never compete with each other.)

I remember listening to IBM's previous CEO, Sam Palmisano, answer a version of the question "What is IBM?" And he said IBM is not a hardware company, not a software company, and not a services company. It is, he said, for lack of a better word, a solutions company. IBM needs all three parts playing their roles in harmony, and IBM needs all three parts constantly evolving. So if that means IBM should not be in the point of sale device business (a recent divestiture) but should be in the enterprise memory storage and "big data" business intelligence engine businesses (two recent acquisitions), then that's what IBM is going to do. (A lot of analysts have missed this, but IBM has bought some "hardware" companies lately. I put the word hardware in quotes because the reality is that it has been a long time since hardware has been purely mechanical or electromechanical. Most "hardware" products are a blend of hardware and software.) Both of the latter hardware-related acquisitions are highly relevant to zEnterprise, in the form of the DB2 Analytics Accelerator as an example.

Anyway, I think analysts that focus only on one or a couple of the "big three" parts of IBM miss the big picture. Fortunately so have IBM's competitors.

by Timothy Sipples January 23, 2013 in Financial
Permalink | Comments (0) | TrackBack (0)

A Newcomer for Dinner

I posted in another journal about a certain unnamed newcomer. (We're talking program here, not a person.) It's a game changer, at least for me and my circle of friends.

The new thingy was a hot topic at a recent dinner. I was out with friends, and discussion eventually landed on this subject.

Dining in Dread

All of as at the table have made a career using "the old thing". We don't reject the new thing, but cannot help notice a lack of important features and function. (See my other post, "Performance Matters ... and why".) So we're more than a little worried. The focus of technology is often hyped. It's tough to see the real value in one product versus another.

Discussing Doozeys

You know the phrase "it's a doozie". It means something that is extreme. Historically, it means something very good, well made, luxurious, excellent. Formally, it's a "Duesy", a Duesenberg, in its day possibly the best car made. (Certainly the most expensive and of unquestioned quality.) They won races and set records and were beatiful. But Duesenbergs were built by hand and could not keep up with cars built via assembly line (Ford, Olds, others).

Gagging on Good Enough

My personal fear is that the majority will use the newer program either ignorantly or having decided that they don't need the better performance and substantial instrumentation of the older program. Sadly, many go the route of "good enough" and get by, ignoring the dollars and sense lost in having to compensate.

Then again, sometimes you just need a Model T. Then again, few of us drive the low end model these days.

Sunset on Solaris

I leave this new thing unnamed, the thing I am bemoaning. It will be outed soon enough. It could be anything. Consider Solaris -vs- Linux. Linux was the newcomer. Solaris was solid. Frankly, Solaris was the better Unix at the start and remained technically superior for many years. (Some would say even now.) Being a Linux aficionado, my statements favoring Solaris are validated. But I can't use Solaris, partly because it is proprietary, partly because the mainframe port was quashed, and partly because I just don't have time.

My hope is that "the new thing" will acquire the most important features of "the old thing". If you cut through the bloat in Linux, and the barnacles attached to it, you see that it has picked up several of the better features of Solaris. We gotta see more of that.

In any case, I ask you, please ... always pursue the best.  Don't settle for "good enough".


-- R; <><











by sirsanta January 11, 2013
Permalink | Comments (0) | TrackBack (0)

The postings on this site are our own and don’t necessarily represent the positions, strategies or opinions of our employers.
© Copyright 2005 the respective authors of the Mainframe Weblog.