scholarly journals High–Performance Computers

1992 ◽  
Vol 10 (6) ◽  
pp. 632-632
Author(s):  
Stuart M. Dambrot
Author(s):  
Jack Dongarra ◽  
Laura Grigori ◽  
Nicholas J. Higham

A number of features of today’s high-performance computers make it challenging to exploit these machines fully for computational science. These include increasing core counts but stagnant clock frequencies; the high cost of data movement; use of accelerators (GPUs, FPGAs, coprocessors), making architectures increasingly heterogeneous; and multi- ple precisions of floating-point arithmetic, including half-precision. Moreover, as well as maximizing speed and accuracy, minimizing energy consumption is an important criterion. New generations of algorithms are needed to tackle these challenges. We discuss some approaches that we can take to develop numerical algorithms for high-performance computational science, with a view to exploiting the next generation of supercomputers. This article is part of a discussion meeting issue ‘Numerical algorithms for high-performance computational science’.


PAMM ◽  
2015 ◽  
Vol 15 (1) ◽  
pp. 495-496 ◽  
Author(s):  
Lennart Schneiders ◽  
Jerry H. Grimmen ◽  
Matthias Meinke ◽  
Wolfgang Schröder

2012 ◽  
Vol 629 ◽  
pp. 704-710
Author(s):  
Xi Ying Liu ◽  
Tong Gui Bai ◽  
Tao Zhang

Analyzing problems represented by partial differential equations numerically with modern high performance computers has become an important approach in research of earth science. In the work, a Sea Ice numerical Model under JASMIN (J parallel Adaptive Structured Mesh applications INfrastructure) (SIMJ for brevity) including thermodynamic and dynamic processes is implemented and an numerical experiment of 20-year integration with SIMJ has been performed. It’s found that the model can reproduce seasonal variation of Arctic sea ice well and implementation of parallel computing is flexible and easy. The ratio of time consumption is 1:1.16:1.48:2.45 with 8, 4, 2, and 1 core(s) respectively for one year integration on mobile workstation (Thinkpad W510) with Red Hat Enterprise Linux 5.4 and Portland group’s pgf90 9.0-1.


Author(s):  
Jackson B. Marcinichen ◽  
John R. Thome

For the next generation of high performance computers, the new challenges are to shorten the distance for transporting data (to accelerate the transfer of information) between multi-microprocessors and memories, and to cool these electronic components despite the increased heat flux that results from increased transistor density. Recent technological advances show a tendency for the development of 3D integrated circuit stacked architectures with interlayer cooling (multi-microchannels in the silicon layers). However, huge challenges exist in such design/concept, i.e. flow distribution to hundreds microchannels distributed in the different interlayers, thermo-hydrodynamic and geometrical limitations, manufacturing etc. 3D-ICs with interlayer cooling are still about a decade away, so a viable shorter term goal is 3D stacks with backside cooling, taking advantage of Si layers now able to be thineer down to only 50 μm thickness. Thus, the present work presents thermo-hydrodynamic simulations for 3D stacks considering only a backside cooler, which simplifies considerably the assembly and guarantees a high level of reliability. In summary, the results showed that this concept is thermally feasible and potentially that interlayer microchannels (between stacks) will not be necessary.


2005 ◽  
Vol 17 (10) ◽  
pp. 1239-1270 ◽  
Author(s):  
Jeffrey S. Vetter ◽  
Bronis R. de Supinski ◽  
Lynn Kissel ◽  
John May ◽  
Sheila Vaidya

Sign in / Sign up

Export Citation Format

Share Document