Saturday, February 28, 2009

How to Build a Super Computer



---->A super computer is a cluster of two or more computers connected via a networking protocol. The cluster is used to distribute task workload between computers and to provide a fail-safe system to keep a high level of system availability. Familiarity with building computers and installing and customizing operating systems, networks and software is required before building a super computer.

---->install the same operating system on the computers to be used for the super computer. Windows NT and Windows 2000 are stable operating system to use for a Windows-based cluster.

Connect the computers for the super computer to a local area network (LAN) hub or switch.

---->Install a cluster management service and communication software on the computer that will serve as the master computer in the cluster. The Message Passing Interface (MPI) is a stable and easy-to-use software for this task. MPICH is a free version of MPI and can be found at the MPICH home page.

---->Extract the software package to a new folder on the main hard drive of your master computer. Copy the ".dll" files from the new folder to the Windows system 32 folder on the main hard drive.

--->Run the ".bat" file for the software to install the cluster management service and communication software.

--->Repeat Step 4 on all other computers in the cluster.

--->Execute the "shell" program found in the new folder on the master computer to detect all other computers in the cluster. Assign each computer a unique address. The super computer is ready to test and use

What is supercomputing?



Seymour Cray at Control Data Corporation introduced this technology, and it was in the market by the 1970s. Supercomputers are computers with huge processing capacity that perform complex amounts of mathematical calculations in a very short time. They are expensive and are used for specialised applications. Some of the spheres in which supercomputers find use are weather forecasting, graphics, animation, electronic design, fluid dynamic calculation, nuclear energy research, petroleum exploration, structural analysis, meteorology and weapon simulation. Supercomputers have traditionally run Unix variants and FORTRAN, the mainstay language of scientific and technical programming.

Which are the notable supercomputers in India?

The two most famous supercomputers developed in India are PARAM and Kabru. The former was the first Indian version of supercomputers to be introduced in 1991 in Pune. PARAM (Parallel Machine) 8000 was first developed by C-DAC (Centre for Development of Advanced Computing). It runs on Linux or IBM’s AIX Operating System, and is capable of making 100 billion calculations per second. India has sold two such machines to Russia and Singapore. The second supercomputer to be developed in India is Kabru; it is also the second-fastest in the country. It is named after Kabru, one of the tallest peaks in the Himalayan range.

Which is the most powerful supercomputer in the world?

Presently, the fastest supercomputer is Blue Gene/L that has 131,072 processors and at its peak executes 367,000 trillion mathematical operations per second (teraflops). It runs CNK/Linux and uses a proprietary inter-connect to hook up a multitude of PowerPC 440 700 MHz processors.

Wednesday, January 16, 2008


Lawrence Berkeley National Laboratory and Cray today announced the successful completion of the acceptance test of powerful Cray XT4™ system contains nearly 20,000 processor cores and has a top processing speed of more than 100 teraflops.

The next-generation supercomputer will be used to advance a broad range of scientific research. Named “Franklin” in honor of the first internationally recognized American scientist, Benjamin Franklin, the Cray XT4 system enables researchers to tackle the most challenging problems in science by conducting more frequent and increasingly detailed simulations and analyses of massive sets of data.

super computers

The twice-yearly TOP500 list of the world's fastest supercomputers, already a closely watched event in the world of high performance computing, is expected to become an even hotter topic of discussion as the latest list shows five new entrants in the Top 10, which includes sites in the United States, Germany, India and Sweden. The 30th edition of the TOP500 list was released today (Nov. 12, 2007) at SC07, the international conference on high performance computing, networking, storage and analysis, in Reno, Nevada.

supercomputers