Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


Archives for April 2006

What is the easiest way to write technical codes?

Computationally intensive codes have traditionally been developed in Fortran or C for the supposed speed improvements. But what customers really want is enhanced productivity, not just for the machine, but for the programmer as well. For this reason, domain-specific languages have really taken off in the technical area, and indeed systems like Excel and even […]

What Is a Co-processor?

Most computers have a central processing unit (CPU) that is capable of general purpose processing, specifically control flow and arithmetic / logic. Applications with heavy processing demands can greatly weigh down the CPU. One solution is to have multiple CPUs, though this can be quite expensive for many cases. Instead, a system designer may add […]

What is Fibre Channel?

Fibre Channel is a storage connectivity; it is serial and can be run over long distance. Fibre Channel’s flexibility is demonstrated in its range of topologies—point-to-point, switched fabric, or arbitrated loop (ring)—as well as its range of supported upper layer protocols, which include IP and SCSI. Its host bus adapter (HBA) plugs into a PCI […]

What is iSER?

As mentioned earlier, SCSI and ATA connect a storage device to a computer. A number of devices now days are no longer strictly internal and may be connected via the bus with an external cable. SAS and SATA both have cables that link an external disk with the machine. In addition, standard general purpose connectors […]

What is the difference between SCSI and ATA?

SCSI (small computer system interface) and ATA (advanced technology attachment) are two standards for connecting storage devices to a computer. Traditionally, ATA was cheaper while SCSI was more powerful, but that simple comparison is no longer valid. Both now feature direct memory access (DMA), which frees the CPU during reads and writes. Both now feature […]

What is a blade?

Traditional clusters are made from 1U-sized compute nodes that are linked together via a network. As explained earlier, this kind of system design suffers from issues with management, fault-tolerance, and power consumption. A blade server attempts to address these concerns by placing minimal compute nodes within a shared chassis. The “blades” are nodes that contain […]

What is MPI?

The Message-Passing Interface is a standard for programming distributed-memory parallel computers. Open source implementations such as Open MPI and MPICH2 are available for many platform, so software developed with MPI tends to be portable. MPI has both one-sided and two-sided message semantics, in addition to collective communication routines. MPI also supports a private context for […]