Architectural Properties for HPC

White Papers > HPC Systems Management > Architectural Properties for HPC

High Performance computer systems can be regarded as the most powerful and flexible research instruments today. They are employed to model phenomena in fields so various as climatology, quantum chemistry, computational medicine, High-Energy Physics and many, many other areas.  In this article we present some of the architectural properties and computer components that make up the present HPC computers and also give an out- look on the systems to come. For even though the speed of computers has increased tremendously over the years (often a doubling in speed every 2 or 3 years), the need for ever faster computers is still there and will not disappear in the forseable future.

Before going on to the descriptions of the machines themselves, it is useful to consider some mechanisms that are or have been used to increase the performance. The hardware structure or architecture determines to a large extent what the possibilities and impossibilities are in speeding up a computer system beyond the performance of a single CPU core. Another important factor that is considered in combination with the hardware is the capability of compilers to generate efficient code to be executed on the given hardware platform. In many cases it is hard to distinguish between hardware and software influences and one has to be careful in the interpretation of results when ascribing certain effects to hardware or software peculiarities or both. In this article we will give most emphasis on the hard- ware architecture. For a description of machines that can be classified as “high-performance” one is referred to.

The rest of the paper is organized as follows: Section 2 discusses the main architectural classification of high-performance computers; Section 3 presents shared-memory vector SIMD machines; Section 4 discusses distributed memory SIMD machines; Section 5 looks at shared-memory MIMD machines; Section 6 overviews the distributed-memory MIMD machines; Section 7 ccNUMA machines which are closely related to shared-memory systems; Section 8 presents clusters; Section 9 overviews processors and looks at what’s currently available today; Section 10 presents computational accelerators, GPUs, and FPGAs; Section 11 discusses networks and what is commercially available; Section 12 overviews recent trends in high- performance computing; Section 13 concludes with an examination of some of the challenges we face in the effective use of high-performance computers.

    Contact Info

    Work Email*
    First Name*
    Last Name*
    Zip/Postal Code*

    Company Info

    Company Size*
    Job Role*

    All information that you supply is protected by our privacy policy. By submitting your information you agree to our Terms of Use.
    * All fields required.