ABSTRACT
For over two decades, supercomputing evolved in a relatively straightforward manner: Supercomputers were assembled out of commodity microprocessors and leveraged their exponential increase in performance, due to Moore's Law. This simple model has been under stress since clock speed stopped growing a decade ago: Increased performance has required a commensurate increase in the number of concurrent threads. The evolution of device technology is likely to be even less favorable in the coming decade: The growth in CMOS performance is nearing its end, and no alternative technology is ready to replace CMOS. The continued shrinking of device size requires increasingly expensive technologies, and may not lead to improvements in cost/performance ratio; at which point, it ceases to make sense for commodity technology. These obstacles need not imply stagnation in supercomputer performance. In the long run, new computing models will come to the rescue. In the short run, more exotic, non-commodity device technologies can provide two or more orders of magnitude improvements in performance. Finally, better hardware and software architectures can significantly increase the efficiency of scientific computing platforms. While continued progress is possible, it will require a significant international research effort and major investments in future large-scale "computational instruments".
- Borkar, S. Designing Reliable Systems from Unreliable Components: the Challenges of Transistor Variability and Degradation. IEEE Micro, 25(6), 10--16, 2005. Google ScholarDigital Library
- Cavin, R., Lugli, P. and Zhirnov, V. Science and Engineering beyond Moore's Law. Proceedings of the IEEE, 100,1720--1749, 2012.Google ScholarCross Ref
- Christensen, C., The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail. Harvard Business Review Press, 2013.Google ScholarDigital Library
- Colwell, R. The Chip Design Game at the End of Moore's Law. Hot Chips, Aug. 2013Google ScholarCross Ref
- Courtland, R. The Status of Moore's Law: It's Complicated. IEEE Spectrum, Oct 2013. http://spectrum.ieee.org/semiconductors/devices/the-status-of-moores-law-its-complicatedGoogle Scholar
- Fuller, S. and Millett, L. (eds.), The Future of Computing Performance: Game Over or Next Level? National Academies Press, 2010. Google ScholarDigital Library
- Heck, S., Kaza, S. and Pinner, D. Creating value in the semiconductor industry. McKinsey&Company, San Francisco, 2011.Google Scholar
- ITRS, International Technology Roadmap for Semiconductors, 2013 Edition: Executive Summary. http://public.itrs.net/Links/2013ITRS/2013Chapters/2013ExecutiveSummary.pdfGoogle Scholar
- Kogge P. and Resnick D. Yearly Update: Exascale Projections for 2013. Technical Report SAND2013--9229, Sandia National Laboratories, 2013. Snir, M., Gropp, W. and Kogge, P., Exascale Research: Preparing for the Post-Moore Era. Technical Report, UIUC, 2011.http://hdl.handle.net/2142/25468Google Scholar
- Tiwari, S. and Theis, N. Recommendations from the NSF Workshop on Interdisciplinary Challenges beyond the Scaling Limits of Moore's Law, Aug. 2010. http://www.nnin.org/sites/default/files/files/NSF_Wksp_August_2010/NSF_BeyondMoore_Report_final.pdfGoogle Scholar
Index Terms
- The future of supercomputing
Recommendations
Experiences with implementing Kokkos’ SYCL backend
IWOCL '24: Proceedings of the 12th International Workshop on OpenCL and SYCLWith the recent diversification of the hardware landscape in the high-performance computing community, performance-portability solutions are becoming more and more important. One of the most popular choices is Kokkos. In this paper, we describe how ...
A Research Retrospective on AMD's Exascale Computing Journey
ISCA '23: Proceedings of the 50th Annual International Symposium on Computer ArchitectureThe pace of advancement of the top-end supercomputers historically followed an exponential curve similar to (and driven in part by) Moore's Law. Shortly after hitting the petaflop mark, the community started looking ahead to the next milestone: ...
Toward performance-portable PETSc for GPU-based exascale systems
AbstractThe Portable Extensible Toolkit for Scientific computation (PETSc) library delivers scalable solvers for nonlinear time-dependent differential and algebraic equations and for numerical optimization. The PETSc design for performance ...
Comments