Technology Spotlight - 2016
Dr. Robert Leland is the Chief Technology Officer of Sandia National Laboratories. He presented the opening talk of the 1st IEEE International Conference on Rebooting Computing (ICRC 2016, http://icrc.ieee.org/), held in San Diego, CA, Oct. 17-19, 2016. Dr. Leland reviewed the history of supercomputing, starting with the mainframe era in the 1950s. This was followed by the vector (parallel) era in the 1970s, the distributed memory (massively parallel) era in the 1990s, and the latest transition to a many-core era with heterogeneous nodes. In each case, US government investment was needed to make the transition from one era to the next. The National Strategic Computing Initiative (NSCI) is the latest effort to promote the development of a new generation of supercomputing hardware and software.
Other presentations from ICRC 2016 are available here.
Prof. Neil Gershenfeld is the director of the MIT Center for Bits and Atoms (http://cba.mit.edu/), the first of a series of “Fab Labs” worldwide, which have pioneered the concept of personalized digital fabrication. In Oct. 2015, he presented a talk at CERN in Geneva, Switzerland, about revolutionizing digital manufacturing, by developing software that merges computation with communication and 3D fabrication.
A recent paper by Prof. Gershenfeld on related topics is given here.
Prof. Gershenfeld also presented a similar keynote talk at the International Conference on Rebooting Computing (ICRC 2016).
Hyperscale Computing refers to large-scale distributed data processing associated with Cloud Computing and Big Data. At the recent Looking Ahead conference in Chicago, Intel VP Jason Waxman spoke about the future of data centers. Waxman projected that within the next decade, due to the rapid growth in cloud computing, the majority of servers in data centers will consist of these scalable, cost-effective systems.
An analysis comparing Hyperscale Computing to High-Performance Computing is available here.
At the recent GPU Technology Conference (GTC 2016), the Nvidia CEO Jen-Hsun Huang explained how next-generation Graphical Processing Units will have major impacts on fields as diverse as artificial intelligence, virtual reality and games, and self-driving cars.
An overview of GTC 2016 is available here.
At a recent Student Leadership Conference of Eta Kappa Nu (HKN), the IEEE Honor Society, Dr. Elie Track, the co-chair of RC, presented an overview of the implications of the end of traditional Moore’s Law scaling, and how IEEE Rebooting Computing is reimagining the entire computing enterprise to continue performance improvements into the future.
For more information on HKN and its various academic chapters, see here.
Hewlett Packard Enterprise is promoting their new Memory-Driven Computer System known as “The Machine”, which they argue will revolutionize computing. Using hundreds of petabytes (~ 1017) of fast universal memory, this will enable more effective access and analysis of Big Data, with greater energy efficiency, speed, security, and convenience.
For further information on “The Machine” see here.
A plenary talk at the recent International Solid State Circuits Conference in San Francisco (ISSCC 2016) was given by William Holt, VP of Intel. In going beyond CMOS, Holt discusses 3D heterogeneous integration and new memory technologies that will enable continuing improvement in processor performance into the future.
For videos of the other plenary talks at ISSCC 2016, see here.
The IEEE Rebooting Computing Initiative has sponsored a video for general audiences, which provides the background and motivation of the Initiative. This features interviews with key members of the RC Committee and leading colleagues in the computing and electronics industries, as well as clips from earlier RC Summits.
A conference on “The Future of Intelligence” was held in Gothenburg, Sweden, in December 2015, sponsored by the Nobel Week Dialogue. The keynote speech was given by Ray Kurzweil, a renowned inventor and computer scientist who is now with Google. Kurzweil focuses on the power of exponential growth in information technology, even beyond Moore’s Law, in fundamentally changing human experience.
Other videos from the conference are available here.
The National Strategic Computing Initiative (NSCI) is a major new initiative of the US government on future computing, focusing on exascale computing. At the recent International Conference on High Performance Computing (SC15) in Austin, Texas, a panel of experts discussed the ongoing implementation of NSCI. This involves a major public-private partnership, including the participation of industry and academia as well as several government agencies.
This presentation is also summarized here.
For further information on SC15, including other presentations, see here.
For further information on NSCI and exascale computing, see here.
Qualcomm has introduced a cognitive computing chip, Zeroth, which enables deep learning in a low-power platform suitable for cell phones. Matt Grob, the CTO of Qualcomm, describes some of the capabilities of this new Neural Processing Unit (NPU) as applied to robot control, image recognition, and Big Data processing.
For further information on Cognitive Computing on the Zeroth Platform, see here
Improved energy efficiency on the device and system level will be essential to continue to enhance computer performance for both small mobile and large-scale systems. The NSF-funded Center for Energy Efficient Electronic Science (E3S) (http://www.e3s-center.org/) , led by UC Berkeley, recently presented a 4th Symposium on Energy Efficient Electronic Systems. The program is available here.
Many of the presentations are available as online videos. Two presentations by RC Committee Members include:
“Millivolt Switches will support better energy-reliability tradeoffs”, Erik DeBenedictis
Watch Video here
“Superconducting computing: Lessons from an emerging technology”, by Scott Holmes
Watch Video here