At this week’s Supercomputing 2008 conference, Microsoft Corp. announced that it had debuted in the top 10 of the world’s most powerful supercomputers. The mega-installation is at Shanghai Supercomputer Center and Dawning Information Industry Co. Ltd., which ranked at No. 10 with 180.6 teraflops, the parallel computing speed, and 77.5 percent efficiency.

Microsoft points out that this is “a truly incredible achievement considering that 12 months ago in Reno, Nevada, Microsoft was at 116 on the Top500 list at Top500.org. This is on the heels of Windows HPC Server 2008 releasing to the manufacturing industry in September.”

Microsoft said it is bringing supercomputing to the masses. “Windows HPC Server 2008 makes supercomputing more accessible to end users by allowing them to harness computing power through a familiar Windows desktop environment,” the vendor said.

Microsoft also said it is offering kits on how to achieving supercomputing capacity on its Windows platform.

Microsoft and Cray Inc. — the grandfather of supercomputer vendors — teamed up in September with an announcement to drive high-productivity computing further into the mainstream with the Cray CX1 supercomputer.

Microsoft also said it is delivering developer tools for Windows-based supercomputing, including Task Parallel Library, Parallel LINQ and Coordination Data Structures for managed code to ease the transition to parallel code. These technologies, along with MPI, MPI.Net and Cluster-SOA, extend parallelism to clusters of thousands of nodes using Windows HPC Server 2008.

IBM may be a big, lumbering multi-billion-dollar corporation that is slow to turn, but boy, when IBMers get imaginative, they get imaginative.

Steve Craggs of Lustratus Research recently marveled on what has become a fast-growing phenomenon: deployments of the open source Linux operating systems on mainframes. “Five or ten years ago, this sort of question would have been unthinkable, but now mainframe users are increasingly facing a choice between whether to use Linux on System z or z/OS to host new mainframe workloads.”

New Linux workloads may be the result of consolidation or initiatives such as SOA, he says.

However, IT managers and specialists appear to be divided as to how effective it is to run Linux on mainframes, versus commodity servers — and vise-versa, whether its better to stick to z/OS, the native mainframe OS. “On the one hand, long-time mainframe guys will say that z/OS has grown up with the mainframe and therefore must be the best choice. But IBM has done a lot to its version of Linux for the mainframe, and Linux bigots will be quick to point out that the license costs will be cheaper and there are strong advantages in standardizing on a portable and flexible operating system enterprise-wide.”

Linux offers flexibility, helps to drive license costs down and leverages widely available skills. The multi-system capabilities of z/OS combined with its close linkage to the System z platform offer the greatest exploitation of System z facilities. “But as always the devil is in the details,” says Craggs. “Worst of all, given the polarized nature of IT in general, the decision makers find it hard to get unbiased advice on such a divisive question.”

Facebook recently announced some milestones, which suggests they are maintaining one heck of a server and storage farm. They announced that their users have now uploaded over 10 billion photos to the site. Since they actually store four image sizes for each uploaded photo, so that’s over 40 billion files.

Facebook also pointed out that 2-3 Terabytes of photos are being uploaded to the site every day, and the companies have just over one petabyte of photo storage. They also noted that they serve over 15 billion photo images per day, and photo traffic now peaks at over 300,000 images served per second.

No doubt their data center folks are a very busy bunch.