Abstract
Hardware changes during the last 30 years reflected the higher and higher integration of chips. Moore’s law in 1965 was the anticipation of a progress of integration for about 10 years. But this process continued and is still ongoing. Different to the past decades the higher chip density is not more accompanied by an increasing processing frequency. Because the power consumption of a chip for a fixed technology is proportional to the power of three of the frequency, it is de facto limited today. Chips exceeding a power dissipation of 100 Watt are not reasonable for everyday computing. With limited frequency we see the progress of integration process going in the multiplication of the number of on chip cores. This might lead to an agglomeration of cores with 80x86/EM64T instruction sets in architectures like Intel’s two cores Woodcrest and quad cores Clovertown or AMD dual core Opteron and Barcelona. Both development lines suggest a moderate evolution of the number of cores with the same total performance increasing rate we saw before. On the other hand we perceive the interest of graphic card vendors like Nvidia and ATI to infiltrate the high performance computing market. This forces the large processor vendors to react in a not anticipated way. AMD bought ATI and will try to exploit the high floating point performance of the graphic card for computing. Beyond their market segment Nvidia tries to establish the new computing platform Tesla. The Sony – IBM –Toshiba cooperation developed the Cell processor introducing a completely new system with unusual specifications. Whereas this machine has been developed for computer games and the High Definition TeleVision (HDTV), researchers discovered the potential of this machine for numerical computing. All these recent developments will force Intel to react for not loosing market shares.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Gebis, J., Patterson, D.: Computer 40(4), 68–75 (2007)
Williams, S., Shalf, J., Oliker, L., Kamil, S., Husbands, P., Yelick, K.: The potential of the cell processor for scientific computing. In: Proc. of the 3rd Conference on Computing Frontiers, Ischia, Italy, CF 2006, May 03 - 05, 2006, pp. 9–20. ACM Press, New York (2006)
Harrison, I., Wonnacott, D.: An empirical study of multiprocessor time skewing. In: Proc. of the Mid-Atlantic Student Workshop on Programming Languages and Systems (MASPLAS 2003), Haverford College, April 26 (2003)
Frigo, M., Leiserson, C.E., Prokop, H., Ramachandran, S.: Cache-Oblivious Algorithms. In: Proc. of the 40th Annual Symposium on Foundations of Computer Science, FOCS., October 17 - 18, 1999, vol. 285. IEEE Computer Society Press, Washington (1999)
Göddeke, D., Becker, C., Turek, S.: Integrating GPUs as fast co-processors into the parallel FE package FEAST. In: Matthias Becker, M., Szczerbicka, H. (eds.) Proc. of the 19th Symposium on Simulation Technique, September 2006, pp. 277–282 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Küster, U., Resch, M.M. (2008). Hardware Development and Impact on Numerical Algorithms. In: Krause, E., Shokin, Y.I., Resch, M., Shokina, N. (eds) Computational Science and High Performance Computing III. Notes on Numerical Fluid Mechanics and Multidisciplinary Design, vol 101. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-69010-8_5
Download citation
DOI: https://doi.org/10.1007/978-3-540-69010-8_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-69008-5
Online ISBN: 978-3-540-69010-8
eBook Packages: EngineeringEngineering (R0)