Present and Future Supercomputer Architectures
نویسنده
چکیده
1941 1 (Floating Point operations / second, Flop/s) 1945 10
منابع مشابه
Beowulf Cluster Design for Physics Simulations: Balancing Current Needs with Future Possibilities
Brems is the Idaho State University Physics Department's Beowulf REsource for Monte Carlo Simulations. Beowulfs present unique engineering challenges because of the many choices in vendors, architectures, and hardware afforded by tapping into commodity server and specialty high performance computing markets. Brems is no exception to this and represents a unique balance of code performance, hard...
متن کاملHigh Performance, Three-Dimensional Bilateral Filtering
Image smoothing is a fundamental operation in computer vision and image processing. This work has two main thrusts: (1) implementation of a bilateral filter suitable for use in smoothing, or denoising, 3D volumetric data; (2) implementation of the 3D bilateral filter in three different parallelization models, along with parallel performance studies on two modern HPC architectures. Our bilateral...
متن کاملPerformance Analysis of FEM Algorithms on GPU and Many-Core Architectures
The roadmaps of the leading supercomputer manufacturers are based on hybrid systems, which consist of a mix of conventional processors and accelerators. This trend is mainly due to the fact that the power consumption cost of the future cpu-only Exascale systems will be unsustainable, thus accelerators such as graphic processing units (GPUs) and many-integrated-core (MIC) will likely be the inte...
متن کاملA component-based framework for radio-astronomical imaging software systems
This paper describes a component-based framework for radio-astronomical imaging software systems. We consider optimal re-use strategies for packages of disparate architectures brought together within a modern component framework. In this practical case study, the legacy codes include both procedural and objectoriented architectures. We consider also the special requirements on scientific compon...
متن کاملState-of-the-art in heterogeneous computing
Node level heterogeneous architectures have become attractive during the last decade for several reasons: compared to traditional symmetric CPUs, they offer high peak performance and are energy and/or cost efficient. With the increase of fine-grained parallelism in high-performance computing, as well as the introduction of parallelism in workstations, there is an acute need for a good overview ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2004