sorted by: recent | see : popular

Content Tagged with infrastructure + microsoft

Updated: To paraphrase (and mangle) StarTrek’s famous tagline: Can Russia be the place where Internet companies boldly go looking for the final frontier of data centers? At least one blog thinks so, and it points to the massive hydroelectric power capacity on tap in Russia. An article in this week’s The Economist points to RusHydro, a Russian company with the capacity to produce 25 gigawatts of electricity.

Much of the unused part is in Russia, RusHydro says. It has 5GW of new capacity under construction and more than 20GW on the drawing board—enough to double production.

Power is seen as the biggest constraint when it comes to building data center capacity. As a way around this conundrum, large consumers of Internet data center capacity have located their facilities closer to energy sources. For instance, Google, Microsoft, and Yahoo have built data centers in Quincy in the state of Washington near a hydroelectric dam where they pay a lot less for power than, say, in Silicon Valley. Google has built a massive facility in The Dalles, Oregon, another location close to power source. (Related stories: The Geography of Internet Infrastructure and Why Google Needs Its Own Nuclear Plant)

From that perspective, it is not so far fetched to imagine that these and other companies could plan on building data centers in Russia. Microsoft has already made its intentions very clear and is planning a data center in Siberia. Google has been slowly expanding its presence in Russia including a recent purchase of Rambler for $140 million. Of course, the big problem is a lack of massive Internet backbone pipes in and out of Russia, but that might be an issue that could be addressed easily.

Why? As we have noted before, there is a lot of capacity being built across the Pacific Ocean. Earlier this week, a new 570 km cable with a capacity of 640 Gbps between Russia and Japan went live.The cable is a joint venture between TransTelecom Company CJSC of Russia, which has about 55,000 kilometers of backhaul network in Russia. The other partner in this cable is NTT. Similarly, Eastern Europe is seeing big build-outs when it comes to fiber to the home (and/or premises). These networks needs backhaul pipes leading to big network upgrades.

I think the reason is that Russia’s natural environment makes it a good candidate for big data center expansion. There are some folks who have come up with ways to leverage natural environments such as cold weather to lower the amount of power required to cool a data center. Andrew Hopper, head of Cambridge University Computing Lab, has been preaching the mantra of putting data centers next to power sources, since it takes “electrical transmissions costs out of the equation.”

Google Data Centers Around The World Map Courtesy of Pingdom. Our source tell us that not all locations on the map qualify as data centers. Instead, some of them are data centers and others are smaller locations that route traffic to “real” data centers.

Virtualization underpins cloud computing by making it possible to separate the software from the hardware. So far the dominant player has been VMware, with about 95 percent of the market, but Simon Crosby, CTO of Citirix and former CTO of open source virtualization company XenSource (which Citrix acquired last year), plans to take that market back. The launch of Microsoft’s Hyper-V hypervisor is part of of his strategy. The rest revolves around services that “play nicely with others,” and free hypervisors embedded into servers and operating systems.

GigaOM: Can you tell me how the launch of Hyper-V affects Citrix Xen products?

Crosby: Our key founding philosophy was fast, free, compatible and ubiquitous hypervisors. Microsoft’s Hyper-V which is compatible with XenServer, is alright when it comes to being fast; it’s 28 bucks, so close to free; and because it’s Microsoft it will be ubiquitous. So for us, it’s good. The problem is it took them too darn long to get it out. Working with Microsoft has been a little bit like having a ring through the nose of the bull. We have a rope tied to that ring because we’re ahead of them on this thing, but when they charge I’m going to get out the way and point them at VMware.

GigaOM: But will Hyper-V compete with the Citrix server virtualization business anyway?

Crosby: You should look forward to interesting announcements of products to add value to Hyper-V. We’re going to sell into that footprint much like Citrix has always extended the use cases of Microsoft products.

GigaOM: What about VMware?

Crosby: We only have a 4 percent or 5 percent share in this market, and the market is significantly overpaying for what they have today, so it’s a very, very interesting time. We’re going to track VMware down with the fast, free, compatible and ubiquitous hypervisor and sell on top of that. We’ve accepted that hypervisors are not the stuff you can charge for. It has taken longer than I thought to get there, and customers have yet to decide, too, if the hypervisors are part of the box or in the operating system. We’ll be wherever we can to create for ourselves the largest possible upsell with other products.

GigaOM: So we’re early on in this game?

Crosby: Virtualization is reorganizing the IT industry. Separating the software from hardware allows more services oriented on the software stack. It creates this huge power vacuum in the industry, and everyone is rushing to fill it. Virtualization becomes a tool for differentiation for a former commodity box maker. Look at our deal with Symantec and its Veritas products. That’s a profoundly important play because, in the enterprise, what you’re about to see is companies entering the virtualization market with a commodity hypervisor and a clear intent to upsell the products.

GigaOM: What will virtualization mean for storage?

Crosby: With Xen, multiple servers will automatically pool, and from these resource pools customers get dynamism and availability. VMware turns the storage industry into a dumb block of boxes, while we have a storage model that allows us to leverage our software to the let the storage infrastructure participate in the value chain, and the storage industry works with us very closely. Microsoft is completely missing from storage.

GigaOM: What about desktop virtualization? Unlike the server side, you guys have a lot of competition ready to pounce.

Crosby: Yes, and that opportunity is of great interest to both of us. Arguably the remote delivery of apps is what we have been doing for 18 years at Citrix. We have always cared about the line of sight between an app in the data center and the desktop. We’re very confident and have opened up the category, but everyone and their dog are in there too. We’re watching 10 to 12 other offerings, but we just see a lot of smoke and not a lot of fire.

GigaOM: After storage, where is virtualization heading next?

Crosby: There’s a lot happening with I/O virtualization and the creation of these fabrics for information flow. Fibre Channel won’t roll over and die, but some of the Ethernet stuff is really interesting. Backing away this thing that has always been proprietary presents interesting opportunities. When you virtualize the resources of a single compute memory you create a new type of system where Xen is the virtualization engine, because it’s not proprietary.

GigaOM: So you’re describing a cloud?

Crosby: To the extent that the clouds are relevant, the largest virtualization effort is Amazon Web Services. Xen will be in every cloud. The only cloud that it won’t be in is Microsoft’s and that will be running Hyper-V. So that’s an interesting path as clouds become an opportunity for some IT functions to be outsourced.

Fresh from his Mix’08 keynote, Microsoft’s Chief Software Architect and industry luminary, Ray Ozzie, spent some time on the phone with me, discussing everything from the company’s services strategy, to the economics of cloud computing, to the relevance of desktop and infrastructure challenges. What follows is a highly edited version of our 20-minute conversation.

Enjoy this interview, the first of what I hope will become a series of conversations with tech greats.

OM MALIK: You outlined Microsoft’s software-plus-services strategy, but what I want to know about is the changing role of the desktop in this service’s future.

RAY OZZIE: I think the real question is (that) if you were going to design an OS today, what would it look like? The OS that we’re using today is kind of in the model of a ’70s or ’80s vintage workstation. It was designed for a LAN, it’s got this great display, and a mouse, and all this stuff, but it’s not inherently designed for the Internet. The Internet is this resource in the back end that you can design things to take advantage of. You can use it to synchronize stuff, and communicate stuff amongst these devices at the edge.

A student today or a web startup, they don’t actually start at the desktop. They start at the web, they start building web solutions, and immediately deploy that to a browser. So from that perspective, what programming models can I give these folks that they can extend that functionality out to the edge? In the cases where they want mobility, where they want a rich dynamic experience as a piece of their solution, how can I make it incremental for them to extend those things, as opposed to learning the desktop world from scratch?

OM: So basically you’re saying that in this new environment, that you have to give up on your legacy of desktop and just view the world from a web perspective?

RAY OZZIE: Well, I can’t say give up on it. Here’s the way I talk about it to people at Microsoft. The desktop is very useful. People use it a lot on a daily basis. There are things that the web is good for, but that doesn’t necessarily mean that for all those things that the desktop is not good anymore. What I think is important is to re-pivot the center of what we are trying to accomplish.

OM: What makes you guys think that you can actually do better than everybody else?

RAY OZZIE: I’m not going to get cocky. The reason I’m a believer is that Microsoft as a company is in a number of different markets. I think we’re well positioned, because we have a selfish need to do these things, and because we have platform genetics. We have the capacity to invest at the levels of infrastructure that are necessary to play in this game. So I think we’ll be well positioned.  I can’t tell you specifically which aspects we’re going to kick somebody else’s butt, or where they’re going to kick our butt, but I think we’re pretty well positioned.

OM: I buy into the whole services model, but then I see what happened a couple of weeks ago — Hotmail goes down for quite a long time. And this happens way too often, not only just at Microsoft, but at other services also. It is very hard for me to imagine, we keep talking about services, but the reliability of the infrastructure is just not there.

RAY OZZIE: It’s not straight engineering, and it’s not an art. It’s somewhere in-between. And we are all learning. And so if you look at the innards of a Yahoo or a Microsoft, an MSN, or a Google, you will see the people who have designed the systems and have taken a number of the things we’ve learned in the enterprise space. We have to throw them them away, because the way that we did it in the enterprise space was more tightly coupled. We need to be more loosely coupled.

So I’m not going to make any excuses for downtime. We need to develop more and better application design patterns that we give to developers that let them develop mesh-oriented apps at birth, horizontal apps that can suffer massive failures of certain aspects of their infrastructure, while still surviving.

OM: It (mesh-oriented apps) sounds like a great idea, but in reality can we actually deliver that kind of a mesh app architecture, and how soon?

RAY OZZIE: I think that you’ll see is over the course of this year, to 18 months, you’ll see the incumbents and startups, both, do their first big volleys of services platform, apps tools, runtimes, various things. It really isn’t being taken seriously right now by anybody except Amazon. They’ve done the world a service by putting out there some fairly provocative, interesting services.

OM: The costs of computing, hardware and bandwidth are dropping quickly. Do you believe that the cost will come down fast enough to make cloud computing actually a profitable business?

RAY OZZIE: Well, it’s unlikely that we would get into it if we didn’t think it was going to be a profitable business. So we’ll just manage it to be profitable. It’s going to have different margins than classic software, or the ad (-supported) business. But, we have every reason to believe that it will be a profitable business. It’s an inevitable business. The higher levels in the app stack require that this infrastructure exists, and the margins are probably going to be higher in the stack than they are down at the bottom.

OM: Can you actually elaborate a little bit on that, like when you say higher in the stack, what precisely do you mean?

RAY OZZIE: Let’s go all the way up. Let’s stick to boring old enterprise, all the way up at business solutions, HR apps, or things like that. Somebody who is selling those apps is going to build in, more than likely, the underlying utility costs within their higher-level service. It will still be cheaper to do those things on a service infrastructure than it is on a server infrastructure, but the margins will still be higher to people who build solutions that customers understand the business value of.

When you go down to selling bandwidth, or selling MIPS there will be competition at that level. So the margins, at generic commodity levels are going to be substantially lower.

OM: When do you think utility computing can be a profitable business; are we’re looking at like maybe two years, four years out before it actually starts to become a profitable entity?

RAY OZZIE: (Let’s) take (one company) who is in the market today: Amazon. They chose a price point. There are either customers at that price point or not. They may have priced themselves at expected costs as opposed to actual today costs, but it doesn’t really matter. They could have brought it out at twice the existing price and there still would have been a customer base, and they’d be making money at birth.

I think all of these utility-computing services, as they’re born will either be breaking even or profitable. At the scale that we’re talking about, nobody can afford, (even Microsoft) can’t afford to do it at a loss. We could subsidize it, I suppose. Google could subsidize it by profits in other parts of their business, we could subsidize it, but I don’t think there’s any reason that any of us in this world would bring out that infrastructure like this without charging for what we’re paying, and then trying to make some profit over it. The cost base is so high in terms of building these data centers you do want to kind of make it up.

• Light Reading: Microsoft Tweaks its Carrier Strategy
• Reuters: Killer Robots Pose Greatest Militant Threat
• Computerworld: What’s Next for 3Com? Another Merger May be in the Works
• InfoWorld: Verizon’s First Open Network Specs Coming March 19
• Telecoms.com: Nokia Unveils Shape-Changing Nano Phone Concept
• ArsTechnica: Spammers Crack Gmail’s CAPTCHA system

Right before the Christmas holidays I got a chance to catch up with Dr. Mendel Rosenblum, VMWare’s chief scientist and one of the company’s four five co-founders. Rosenblum is also an associate professor of computer science at Stanford University, where he leads a group focused on operating systems research. It was at Stanford where Rosenblum and three of his graduate students — Scott Devine, Edouard Bugnion and Dr. Edward Wang — came up with the idea that led to VMWare (VMW). Diane Greene joined them as CEO and the fifth co-founder and the company went public in August, garnering a multibillion-dollar valuation that triggered a virtualization frenzy.

Given that VMWare was in a quiet period prior to the release of its quarterly results, my conversation with Rosenblum was quite general. But he did share with me, among other things, the story of how VMWare got started and his outlook for virtualization in 2008. Here are excerpts from the interview:

How did VMWare get started?

I was a professor at Stanford University and we were building a supercomputer called the Flash Machine. I didn’t want to crunch numbers on this machine, but wanted to use virtualization to see if we could run commodity OSes on [it]. We could, and we wrote a paper about it, and that generated a lot of interest, including from Microsoft, who emailed us and wanted us to come and present to them in Redmond. My grad students who worked with me on the project thought we could commercialize the technology, and in 1998 we launched VMWare.

What was the plan when you launched it?

Clearly, the technology was going to be hard to commercialize, and we decided to focus on doing virtualization on the desktop. We worked on the technology and my wife took care of the business side of things.

It seemed to have been a long time in the making.

It took a lot longer than I thought it would take. We released it first on the Linux platform, because we felt the Linux community would adopt it much faster. That proved to be a good move.

Funny now that you have proved it, there is competition coming out of the woodwork. Oracle and Microsoft, for example.

VMWare clearly is going to have competition. Sure it was nice when we were all alone, but we are very different from these other companies. Oracle and Microsoft, for example, are focusing on single machines for now. That’s a nice thing to do. We used to do that. It is good for server consolidation and it is easier and simpler.

What we are doing is basically coming up with a new way to run the data center. So from that perspective, we will continue to have something better than others.

Let’s talk about the data center for a minute. Do you think the whole architecture of the data center needs rethinking?

We went down a rat hole on how we built the data centers. I am not surprised with all the problems we are having with data centers. In my opinion, the architecture has problems because it was built with inferior solutions. What you had was people placing services on servers in a way that led to lightly loaded machines that were idle most of the time. The whole thing was built for peak performance (and not maximum utilization.) Well, idle machines use as much energy as fully utilized machines. The way out of this is to put more on the machines, and get them to be more efficient and take on the work load that will, to some extent, lower the power consumption.

I wrote about pizza boxes becoming a problem, mostly due to low utilization and higher power consumption. It kind of ties in with your thesis.

You have to see them not as boxes but as resources. People are now beginning to utilize virtualization and federate these pizza-box servers. I think if you start to view them as one unit, you can get more utilization out of them. I think in coming months you are going to see a big push to make all servers (and other hardware) inside a data center look more like a single unit. Ironically, if you look at the future — low-end pizza box servers with multicore CPUs running our software — you will start to see the big machine we were building where we got started.

What is your forecast for 2008 from a virtualization standpoint?

We are in a transition period. I think a lot of people dipped their toes in virtualization and got started with server consolidation. They bought into it the “money-saving” argument. In 2008, I expect people to fully embrace virtualization and extend it to other parts of their businesses, even bringing it in-house and using it for optimizing their desktop infrastructure. More importantly, you will start to see the long-term impact of virtualization in the next 12 months.