I bought the December issue of Wired magazine because it had a feature on Ray Ozzie, the inventor of Lotus Notes and Groove. Lotus Notes is an interesting technology that is one of the best enterprise collaboration tools available today. It seems that the major global “knowledge enterprises” pick Lotus Notes as their collaboration infrastructure. Some of the Big Four (PwC, Deloitte, KPMG and E&Y) use Lotus Notes. And a market leader in the management consulting industry: McKinsey & Co uses Lotus Notes.
Ray Ozzie, as the new Chief Software Architect, envisions Microsoft leveraging it’s market power in the new “Cloud Computing” market. And they plan to do this with a 4-prong strategy. After the break, we look at these 4 strategies and talk about how the open source world is going to respond.
Firstly, Windows Azure. This is roughly similar to Amazon’s EC2 cloud service offering and Aptana’s cloud offering, with a twist: being tied into the MS platform. Although support for Java and LAMP is promised in the future, “From a customer standpoint [Windows Azure] is a very limited offering,” says Chu. “It forces a new proprietary architecture on customers, it only works if all of your apps are based on .NET, and you’re then dependent on Microsoft as your sole service provider for your mission critical apps.” A Gartner blog adds: “But, Azure doesn’t support existing applications – applications need to be targeted originally for Azure. With Azure, sourcing is not a runtime operations decision, it is an application design decision. The software and/or services decision is kinda hardwired at design time, which is unfortunate, and means that enterprises will need to look elsewhere for solutions to cloudsource some of their computing requirements”.
The appeal of one-click deploy from Windows desktops into Windows Azure from within Visual Studio will be a very big selling point for Windows Azure. You can experience this today with Aptana’s Cloud offering, with a pure open source stack (pick between PHP or RoR, with MySQL goodness), and freedom to choose your hosting provider without being subject to proprietary vendor lock-in. And we have still yet to see all of Sun’s cards in this grand poker game.
Secondly, Zurich. This is billed as a set of tools for developers to manage their own cloud-based services. The Open Source world has a similar tool in the Enomalism platform, and Sun’s xVM portfolio. These are essentially tools for monitoring instances of virtualised OSes. The key selling point here is “cloudbursting” (a term coined by blogger/CTO, Reuven Cohen) which means that using a uniform cloud interface, workloads can be shared both within your enterprise cloud, and outsourced to any number of outside providers of cloud capacity. This promise of turning previously high fixed costs (design, build and operate your own enterprise data centre, staffing and training costs, license fees etc) into variable costs that can elastically accomodate large spikes of demand will be the panacea that enterprise and government CIOs long for. Imagine running a sparse data centre, that caters to “normal volume”, but that has the ability of “insourcing” spare capacity out to other related businesses, and that can also “outsource” surges in demand to outside providers.
Thirdly, Live Mesh. Of course, one of the largest consumers of Microsoft’s cloud infrastructure will be … Microsoft. In a case of “eating your own dog food“, one of the biggest demos of Azure will be Microsoft’s copy of Apple’s Mobile Me. iPhone users will be familiar with Apple’s web-based SAAS that makes it easy to sync calendars / pix / emails between your Macs, your iPhone and the cloud.
Fourthly, Office Web Apps, which seems to be a direct competitor to Google’s Docs service. After Ballmer disses Google’s approach to cloud-based SAAS apps, Office Web Apps seems to be copying the market leader. On the downside, MS want to offer Office Web Apps for free, on an ad-supported basis, or on a premium pay basis. Well, if Hotmail/Windows Live is any indication of their performance, I’ll stick to Google thanks very much.
The bottom line
Well, the Open Source world already has technology similar to what the Wired article is heralding as “disruptive technology” on Microsoft’s campus. The GNOME project has the “Online Desktop” which could use a little bit of polish.
It seems that Microsoft has the upper hand re: a single coherent vision. Yes, you can integrate various OSS into your cloud, and base it on an open provider’s technology. But the allure of Azure will be that “all-in-one” integrated feeling of getting everything (OS, development environment, cloud services, etc) from a single vendor.
Players like IBM, Sun Microsystems, Red Hat, Google and Amazon all have an opportunity to change the rules of the game – but in this grand game of poker only Microsoft have really hinted their cards, while Google and Amazon have their market share at stake in the pot. For example, Fedora 10 already has remote installation of virtual machines, but we don’t yet have a single-click cross-platform mechanism to deploy Fedora-based VM’s to a cloud.
The potential upside for the winner in this game is nearly unquantifiable. Computing over the next 3-5 years will be dominated by cloud computing, virtualisation and SaaS, and an early win now could very well spell market power for some years to come.
No related posts.