Enterprise Unix Roundup: Unix Heads for the Clouds

By ServerWatch Staff (Send Email)
Posted Feb 27, 2008


Brian Proffitt
The recent fight over the OOXML format as a standard could be a big clue for determining the new direction for IT. A direction that could leave all of the Unix operating systems sitting pretty.

Over in Geneva, Switzerland this week, the International Standards Organization is getting together for a Ballot Resolution Meeting to determine if Microsoft's Office Open XML (OOXML) format is to be declared a genuine, bonafide international standard. The open source community, which prefers the Open Document Format (ODF) because it is supported in several open source office suite applications, most importantly the OpenOffice.org suite, is rather upset by the whole notion of another format being declared a standard. It does not help that 1) Microsoft proposed the standard, and 2) the format itself had been described my many third-party observers as very complicated.

Microsoft, for its part, has been extremely active in getting this format approved by the ISO. The why for this seems obvious: make your product a standard, and it's easier to sell. That, though, is more of a short-term goal. Long-term, I believe, it has another goal in mind: a paradigm shift in IT that could change the traditional PC/server approach into something we are just now seeing the glimmers of today.

This shift could explain a lot about the corporate movings and shakings in the Linux and Unix worlds. Perhaps all the way back to IBM's entry into Linux back in 2000. More recently, this shift in priorities and methodology could be the real reason why MySQL AB was just purchased by Sun Microsystems for $1 billion.

Enterprise Unix Roundup

Back in October of 2007, Robert Cringely noted something very curious going on at Google. It seems he noted that Google was putting many code extensions into the MySQL database product. This showed up when MySQL had laid out its development roadmap back in the fall. Cringely mentioned it because although Google is MySQL's largest customer, it had never contributed code on such a scale before.

The extensions, Cringely went on to explain, were specifically designed to assist Google Apps to run more efficiently:

While Google has long been able to mess with the MySQL code in ITS machines, it hasn't been able to mess with the code in YOUR machine and now it wants to do exactly that. The reason it will take so long to roll out MySQL 6.1 is that Google will only deliver its MySQL extensions for Linux, leaving MySQL AB the job of porting that code to the 15 other operating systems it supports. That's what will take until early 2009.

At the same time, he noted, Google and IBM were starting to heavily promote the concept that, if it reaches it's full potential, would change IT dramatically: cloud computing.

Cloud computing, it should be noted, is basically the newest iteration of what used to be called utility computing, which in turn used to be called grid computing. The idea of cloud computing is that you can run any application on a remote machine or machines and treat it as if it were installed locally. This differs from grid computing, which really had to have apps optimized for the grid to really work, and utility computing, which also has apps installed remotely, but without as much customization. Cloud computing, at its fullest potential, means you get to use any application as if it were installed locally — except that it's really hosted on a ridiculously fast server farm — "the cloud" — somewhere.

For example, Amazon Web Services, a division of the book/toys/everything seller that was started in 2002, just launched its SimpleDB service in December, after having already launched its Amazon Simple Storage Service and Amazon Elastic Compute Cloud products. These three apps (with more to come, no doubt) can be run through a Web interface and deliver huge computing power to individuals and small companies on an as-needed basis. That SimpleDB database's description even highlights the key thing the underlies all of the cloud computing concept: "Amazon SimpleDB is a web service for running queries on structured data in real time."

"Structured data" — what's a good way to contain such data? In a well-built structured data file format of course. Like, for example, ODF. And who has a vested interest in ODF? IBM certainly does. So does Sun. These two companies — along with Google, Microsoft, and many others — realize that if cloud computing does indeed take off, it will be the file format that makes the whole thing work.

Which is why Microsoft feels it must get its format standardized. Even with tactics that ironically have started to attract the attention of the EU, again. How else can it get a piece of the cloud pie?

After all, Microsoft doesn't make hardware the way IBM and Sun do, and its operating system on the server side is being matched and surpassed by a host of free operating systems (e.g., Linux, BSD and OpenSolaris), as well as the Unix family. It doesn't own the wired network (governments and telecom companies do), and it doesn't have a foothold in the wireless spectrum (but Google will soon). It doesn't have a portal with a huge amount of popularity (though a successful Yahoo! acquisition would solve that problem — something another analyst noted). And on the database side? Well, let's see, didn't I already mention a certain open source database company that Sun recently bought for $1 billion?

And if — if — cloud computing advances to the point where we do indeed use devices that connect to the cloud with virtually no on-board software, Microsoft will be shut out of that part of the cloud, too. Its embedded OS has not demonstrated a real market strength or scalability. But Linux has.

This means, as far as I can see, that Microsoft has few options left wide open to it. Competitors and competing technology have closed a lot of doors on the path to cloud computing success. It can still shove its way through those doors, but that's a much more difficult proposition than getting a solid position in one part of the cloud concept before anyone else does. To use another analogy, it's like a game of musical chairs, with the ISO standard chair being the last empty chair left and many companies trying to beat Microsoft to it.

After a little reasearch on the 'Net, it became clear to me that I was certainly not the first to think of this notion of the cloud as the ultimate end goal for IT. Other pundits have also speculated that the Sun/MySQL purchase was part of a Sun strategy to maintain a place at the cloud table, along with Google. So I cannot take credit for original thinking here.

Reading though all the material about IBM's very big push for cloud computing (and grid and utility computing before that), I did get to pondering: What if IBM's investment in Linux back in 2000 had this kind of scenario in mind? A free operating system with tremendous growth, excelled scalability (up and down) — it even came with a network-transparent GUI and a built in commnunity of developers. Perfect for a long-term plan to get into grid/utility/cloud computing.

I have always wondered why IBM didn't put just a little effort into desktop Linux/Unix development. Too divergent from its core AIX and mainframe business? Maybe ... or maybe it realized that the desktop as we know it could be a technological dead end.

It would certainly lend credence to why IBM has such a vested interest in seeing OOXML fail. A chance to shut its old enemy Microsoft out of what could be the next step in IT infrastructure? Hardly a tear would be shed in Armonk on that day, I'm sure.

If OOXML does not become an ISO standard, Microsoft would lose the format war for the cloud, and any advantages SharePoint might have in the face of similar systems like Alfresco's. There would be no reason for vendors to be forced deal with Windows or WinCE on the device end of the cloud, either.

And out on the servers and mainframes that will be serving up this cloud goodness? Unix, BSD, Linux ... all secure, all ready to scale for any sized job.

Pretty scary ... for Microsoft.

Brian Proffitt is managing editor of JupiterWeb's Linux/Open Source channel, which includes Linux Today, LinuxPlanet, and AllLinuxDevices.

Page 1 of 1


Comment and Contribute

Your name/nickname

Your email

(Maximum characters: 1200). You have characters left.


 

 


Discuss this article in the ServerWatch discussion forum

Unsure About an Acronym or Term?
Search the ServerWatch Glossary