By Mary Jo Foley
February 1 is considered the “one year” anniversary of Microsoft’s Azure cloud platform (even though February 2 is the actual date that billing was “turned on”).
As I noted last year, Microsoft has been slowly and steadily adding new features to Azure. But I haven’t written much about longer-term Azure futures. Until today.
Bill Hilf, General Manager of the Technical Computing Group (TCG) at Microsoft, isn’t part of the Azure team. But he and his band are doing work on technologies that ultimately may have substantial bearing on the future of Microsoft’s cloud platform. The TCG has a server operating system team, a parallelization team and a team “with the idea of connecting a consumer to a cloud service,” according to Hilf.
The TCG late last year stated its intentions to allow customers to provision and manage Windows Server 2008 R2 HPC nodes in Windows Azure from within on-premises server clusters as part of Service Pack 1 of HPC Server 2008 R2. But Hilf and his team want to go beyond this and turn the cloud into a supercomputer, as Hilf explained to me last week. “We want to take HPC out of niche access,” he said.
This isn’t going to happen overnight, even though the biggest Azure customers today are the ones using HPC on-premises at the current time, Hilf said. HPC and “media” (like the rendering done by customers like Pixar) are currently the biggest workloads for the cloud, Hilf said.
To bridge HPC and Azure, Hilf has a multi-pronged strategy in mind. One of the prongs is Dryad.
Dryad is Microsoft’s competitor to Google MapReduce and Apache Hadoop. In the early phase of its existence, Dryad was a Microsoft Research project dedicated to developing ways to write parallel and distributed programs that can scale from small clusters to large datacenters. Both the Bing and the Xbox Live teams have used Dryad in building their back-end datacenters.
Dryad is Microsoft’s competitor to Google MapReduce and Apache Hadoop. In the early phase of its existence, Dryad was a Microsoft Research project dedicated to developing ways to write parallel and distributed programs that can scale from small clusters to large datacenters. Both the Bing and the Xbox Live teams have used Dryad in building their back-end datacenters.
There’s a DryadLINQ compiler and runtime that is related to the project. Microsoft released builds of Dryad and DryadLINQ code to academics for noncommercial use in the summer 2009. Microsoft moved Dryad from its research to its Technical Computing Group this year.
“Dryad, in its first iteration, is really for on-premises,” Hilf told me during an interview last week. “Eventually, we’ll roll Dryad up into Azure, as even more data is put in the cloud.”
Before that happens, Microsoft will include Dryad in Service Pack 2 of HPC Server 2008 R2, Hilf said. He didn’t have a firm ship-target, but hinted “late 2011″ might not be a farfetched timeframe. (Microsoft announced last week that it also is making the Dryad code available to interested testers as one of the technologies in its newly formed TC Labs subsection of DevLabs. (TC Labs will be for “any software that is not yet in a product, and of which all or none of it might become part of a product” at some point, Hilf explained.)
Another TCG prong which could have impact on Azure’s future is what Hilf is calling the “parallelization stack.” This stack includes the runtimes, languages and other parallel/multicore tools and technolgies that Microsoft has been building for the past couple of years. This parallel-computing layer/platform will be rebranded around the time of the Professional Developers Conference, Hilf said (which I hear is happening in the early fall of 2011).
It’s the combination of these technologies — including Dryad and the parallel stack — which will enable Microsoft to create an abstraction layer that will allow users to access compute resources — whether they’re on multicore PCs, servers and/or the cloud, Hilf said. The customers most likely to benefit are the real data wonks — the folks that Hilf calls “domain specialists.” These are the individuals in financial services, manufacturing, oil and gas, media, the hard sciences, and other data-intensive professions who have an insatiable appetite for data.
It’s still early days for this kind of supercomputer-in-the-sky stuff. But Hilf already is talking up things like a technical-computing PaaS (platform as a service) and TC SaaS (software as a service).
I’ve got to say that Hilf’s whiteboarding got me wondering whether Azure will remain Microsoft’s core cloud PaaS in the longer term, or if something like the still-mysterious Midori operating system could come into play here. As the M-word (Midori) is not allowed to be uttered in the hallowed Redmond halls by us bloggers/journalists, I didn’t even attempt to ask. I did ask about Orleans — a future cloud programming model technology that’s being pioneered in the more research-centric Microsoft eXtreme Computing Group — and was told by Hilf that it looked interesting in concept, but that it was too early to say if/how/when it could come into play.
My biggest take-away from my interview with Hilf is Microsoft isn’t waiting around for its business customers to overcome their cloud objections. Microsoft is looking for other ways to attract enterprise customers to the cloud, with nearly unlimited data-access and processing power as the lure.
http://www.zdnet.com/blog/microsoft/windows-azure-futures-turning-the-cloud-into-a-supercomputer/8592?alertspromo=&tag=nl.rSINGLE
http://www.zdnet.com/blog/microsoft/windows-azure-futures-turning-the-cloud-into-a-supercomputer/8592?alertspromo=&tag=nl.rSINGLE
