ACM, IEEE Call Dally a Visionary for his Work Advancing Parallel Processing
"This wonderful recognition reflects how Bill's pioneering work in parallel
processing is on its way to revolutionizing computing," says Jen-Hsun Huang,
NVIDIA CEO and president. "We are delighted to have the benefits of his
singular talent as we endeavor through our GPUs to bring parallel computing
to the world."
Previous winners of the Eckert-Mauchly Award include Seymour Cray, a key
figure in the birth of supercomputing; David Patterson, a computer pioneer
teaching at University of California, Berkeley; and Stanford president John
In recognizing Dally for his achievements, the ACM and IEEE wrote: "Early in
his career, Dally recognized the limitations of serial or sequential
processing to cope with the increasing need for processing power in order to
solve complex computational problems. He perceived the ability of parallel
processing, in which many processing cores, each optimized for efficiency,
can work together to solve a problem."
Parallel processing has expanded in recent years from its traditional realm
of environmental science, biotechnology and genetics to applications in such
areas as data mining, oil exploration, Web search engines, medical imaging
and diagnosis, pharmaceutical design, and financial and economic modeling.
NVIDIA's TeslaTM graphics processing units and its CUDA architecture are
key tools enabling this transition.
The organizations note: "Dally developed the system and network
architecture, signaling, routing, and synchronization technology that is
found in most large parallel computers today. He also introduced the Imagine
processor, which employs stream processing architecture, providing
high-performance computing with power, speed, and efficiency."
Dally will receive the 2010 Eckert-Mauchly Award at the International
Symposium on Computer Architecture which will take place June 19-23, in
Prior to joining NVIDIA last year as chief scientist, Dally served from 2005
to 2009 as chairman of Stanford¹s Computer Science department, where he had
been a computer science professor since 1997. Previously, he led the group
at Massachusetts Institute of Technology that built the J-Machine and
M-Machine, parallel machines which pioneered the separation of mechanism
from programming models. Previously at the California Institute of
Technology (Caltech), he designed the MOSSIM Simulation Engine to provide
the computing power required to verify complex Very Large Scale Integration
chips. He also designed the Torus Routing chip, a self-timed chip that
reduces the latency of communications that traverse more than one channel.
Dally has published more than 200 papers and holds over 75 patents. He is
the author of two textbooks, Digital Systems Engineering andPrinciples and
Practices of Interconnection Networks. A Member of the National Academy of
Engineering, a Fellow of ACM, IEEE, and the American Academy of Arts and
Sciences, he received the 2000 ACM Maurice Wilkes award and the 2004 IEEE
Computer Society Seymour Cray Computer Engineering Award.
Dally received a B.S. degree from the Virginia Institute of Technology and
an M.S. from Stanford, both in electrical engineering. He has a Ph.D. in
computer science from Caltech.