Amazon Cloud Computing Support

Amazon continues to roll out cloud computing offerings at a blistering rate.  Today, they just announced a toolkit for Eclipse, the open source IDE.  I’ve been playing around a bit with S3, their storage service, and EC2, their virtual server offering (although I wish that they would offer SLES in addition to OpenSUSE.)  They also have a database and a content distribution network.  Not bad for a bookstore.

I recently talked to a friend about building out their data center and my immediate response was, “Why on earth would you want to build a data center?”  Of course, there are still good reasons but there are fewer and fewer of them each day.

Continue reading

Head in the clouds

Michael Nygard has his head in the computing clouds, suggesting that not only is cloud computing in our future, but that there’ll be many of them. He’s right.

Everyone who runs a large data center is today faced with the same set of interconnected environmental problems; space, power, and heating/cooling. And these are environmental not just in the sense of tree-hugging but also in a straightforward practical sense: there is no more space, there is no more power, there is too much heat and not enough cooling. These problems were the domain of junior people a few years ago, worrying about where, physically, to locate all the new Windows boxes. Then it was middle managers trying to sort out power and HVAC issues: “If we deploy a new phone system in our building we won’t have enough power to do any upgrades in the data center,” that sort of thing. Now environmental issues are front-and-center for senior IT management and if you’re a “red-shift” kind of company, for senior corporate leadership too.

You can cloak it if you want to in green terms but businesses are faced with real operational issues that they need to address regardless of their perspective on global warming or riverine dolphins.

Alongside these environmental issues, data centers are also facing a crisis of manageability. A large enterprise data center is a staggeringly complex thing, too complicated. Also, if the truth be told, most of them are not that well run; would you expect, for example, that an auto parts distributor would have great technology management skills? No, of course not, and the fact is that they probably wouldn’t want to spend the money to acquire that talent and technology even in they could; their differentiation, the competitive advantage of their business, lies elsewhere. So they have a complicated, and sub-optimized, technology infrastructure.

The answer to all of these problems — Monday edition — supposedly lies in virtualization. Novell gets brought into these conversations because inevitably data center managers have a roadmap that looks something like this:

Continue reading

Conservative IT

Enterprise IT, and the people that run it, are risk-averse. Things that work are valued, highly, over new things. The kids might all be learning Ruby and Scheme, but COBOL and C/C++ still rule in the enterprise, where Java is seen as an up-and-comer. Think mainframes are old news? Then you haven’t spent a lot of time in an enterprise data center.

I spoke recently with a guy worked at a VMS help desk twenty five years ago; he said that he’d recently run into some old colleagues from that time and asked them what they were doing. They said they were doing the same thing, VMS support, and that the team had pretty much stayed the same size, a couple of dozen people. IBM supposedly still has their own VMS help desk for their internal users. (You will recall that VMS is an old DEC operating system, an ancient enemy of IBM’s, so this is an admission not only that they use a competitor’s operating system but also, more to the point, that they can’t get off of it.)

Continue reading

A Yahoo smaller

The MS-YHOO deal is keeping the merger arb guys up at night.

Michael Arrington notes that the Yahoo acquisition is getting very expensive, in terms of Microsoft’s market cap; Microsoft has “lost nearly $40 billion in market cap in the eight trading days since they made their offer.” In other words, “Microsoft has shrunk by a Yahoo in the last eight days.”

Henry Blodget has a must-read piece on the logic of the deal; he argues that Microsoft is confusing the ad-driven consumer business with the license-driven corporate business. This makes sense to me; Microsoft needs to defend the Office franchise in the corporate environment, where even Google is getting license revenue ($50/user/year for Google Apps) in lieu of advertising. Enterprises aren’t going to use ad-supported free software, at least not in any future I can see. So why should Microsoft take on the pain that is Yahoo for the consumer side? Blodget writes, “Put differently, the part of Google that threatens Microsoft’s core Windows and Office business is Google Apps, not Google Search.”


A former client of mine, an enterprise architect and a guy I really respect, recommended Splunk (“not just a dirty word”) to me. They bring, more or less, a search engine approach to log file analysis. Now, this is not the sexiest thing in the world, but it’s critically important, especially in large IT shops. A large enterprise generates humongous amounts of log files; my friend said that he’s pinned a big server with just the logs from their domain controllers. And remember, these are just text files.

So the question becomes: how do you analyze all this? Traditionally, people have taken a static reporting approach, which has its place, but you need more when you have to be actively responsive. When was the last time David Hasselhof logged on? Where was he? What systems did he log onto? Did he look at Michael Jackson’s billing records?

Long ago, people thought that some kind of library-like structure was required in order to discover information on the Internet, but it turned out that brute-force searching was better. Likewise in this case, where the end goal is a Google-like interface. Now, this approach has its limitations. You have to know what you’re looking for, first of all. It doesn’t do correlations. It’s got a beautifully simple interface, but it’s not an easy UI for normal, proactive review. It’s not for canned reports. It’s not a SEM (Security Event Manager, or SIEM: Security Information and Event Manager) tool.

But for what it is, it’s great. It’s easy to look at Splunk and say, “you’re just indexing text,” but there is great power in that; look at Google. There been such a huge emphasis on auditability that we’ve generated huge files of events, but mostly they just sit there unloved. Splunk is a good way to leverage that resource.