Microsoft Finally up for Distributed Computing?

ReeprFlame writes "eWeek has reported overhearing Microsoft's plans to finally get into the distributed computing market. Considering that the Windows platform has never had the ability to parallel compute in the past, it leaves great potential to the company's operating system development. From current *nix systems we have today, such a grid proves very useful, especially in the serving arena. However, we are unsure of Microsoft's target for the software. Would it be an addition to home users computers as well as the server versions of Windows? As of now it is unclear, but Microsoft probably will bring this situation to life in the near future since it does hold alot of power for them over other platforms."

Comments

confusing parallel and distributed computing

[goombah99]

the article poster seems to confuse parallel processing on a single machine with distributed computing. The difference is that each machine is running it's own OS and not sharing physical memory in distributed computing.

distributed computing happens at the application layer. Thus if you can run something like an MPI library on windows you have the basis for efficient distributed computing. All you need is a scheduler and launcher to be able to launch distributed launch an application across the net. But virtually all of these are daemons not strictly part of the OS. So that level of system independent abstraction exists already so this should not be too difficult.

Re:Who wrote the summary?

[Scott7477]

Here is tha actual article; note that MS doesn't plan to have this ready to release until "near the end of the decade."

A Peek Under Microsoft's Secret 'Bigtop'
By Mary Jo Foley, Microsoft Watch
December 29, 2004

Microsoft officials have said little about the company's intentions in the grid-computing space. But that doesn't mean Microsoft is ignoring the evolving arena of grid/distributed computing.

Microsoft is working on a skunk-works project that is code-named Bigtop, which is designed to allow developers to create a set of loosely coupled, distributed operating-systems components in a relatively rapid way, according to sources close to the company, who requested anonymity.

Rather than attempting to tightly couple a few high-performance systems together, Microsoft is looking at the consequences of loosely coupling a larger number of moderately powerful computers to achieve a similar result.

Bigtop's first commercial manifestation will likely be as some kind of large-scale project, most likely a distributed grid-computing operating system, the sources added.

Bigtop is one of Microsoft's incubation projects. It falls under the domain of Craig Mundie, the Microsoft senior vice president and chief technical officer in charge of advanced strategies and policy, sources said.

Bigtop consists of three components, all written in C#, according to developers who said they were briefed by Microsoft. These are:

Highwire: Highwire is a technology designed to automate the development of highly parallel applications that distribute work over distributed resources, the aforementioned sources said. Highwire is a programming language/model that will aim to make the testing and compiling of such parallel programs much simpler and more reliable.

Bigparts: Bigparts is code designed to turn inexpensive PC devices into special-purpose servers, according to the sources. Bigparts will enable real-time, device-specific software to be moved off a PC, and instead be managed centrally via some Web services-like model.

Bigwin: According to sources close to Microsoft, Bigwin sounds like the ultimate manifestation of Microsoft's "software as a service" mantra. In a Bigwin world, applications are just collections of OS services that adhere to certain "behavioral contracts." These OS services can be provided directly by the core OS or even obtained from libraries outside of the core OS.

Sources said Microsoft will likely make some sort of preview version of the Bigtop code available to the company's software-development partners by 2006. If and when the final version debuts, it won't be much before the end of the decade, sources added.

It's not clear whether the Bigtop components will run on top of Windows when they are completed. But sources say that is what they are expecting at this point. End of Article

I like their use of a circus term as a name for this project. It gives the impression of a bunch of clowns running around into each other and falling down. Kind of like MS systems on the web now.

Re:Windows clusters don't make sense

[Anonymous Coward]

Processing nodes usually don't even have a monitor or keyboard, much less a GUI and a mouse. Windows isn't much use there.

We have some 2500 Windows servers where I work. None of them have monitors, keyboards, or mice. If we need a KVM it's typically to get into the BIOS, not the operating system.

Nor can you strip out the parts you don't need, or customize the kernel for performance.

You most certainly can do both. It costs money, of course, but remember that we're not talking about trivial tweaks like compiling the kernel for your particular processor family. We're talking about hiring a team of programmers to extensively customize the kernel so it runs your specific application and nothing else. That costs a bucket of money, and compared to that the cost of a Windows source code license is not going to be a whole lot.

I still feel that Linux would be a good bit cheaper, but we're talking big bucks both ways. And it's also worth mentioning that Microsoft's licensing model for "corner cases" like this is extremely flexible: they may give the source away at a significant discount just for the publicity. They've done it plenty of times before. Some of those 2500 servers at work run a custom-built NT kernel and we sure aren't a huge international company.