|
Agency contact Bob Decker Redpines +1 415 409-0233 |
EEMBC contact Peter Torelli, President and CTO EEMBC +1 (203) 423-3179 |
EEMBC Recognizes 17 Members for
Contributions to Industry-Standard Processor Benchmarks in 2017
Engineers Honored for Leadership in
Working Groups for Internet of Things, Autonomous Driving, Internet Security, Machine
Learning, and Low-Power Processor Benchmarks
PORTLAND,
Ore. — Aug. 8, 2018 — EEMBC, an industry consortium that develops
benchmarks for embedded hardware and software, today honored 17 members whose
contributions over the last year have gone above and beyond in making the
organization the industry’s standard bearer for benchmarking. The honorees are
being recognized for their exceptional efforts in working groups developing
benchmarks for internet of things (IoT), autonomous driving (ADAS), internet
security, machine learning, and low-power processors, and for their
contributions to the greater good of the industry.
“The
success of EEMBC and its impact on the industry hinges on the volunteer efforts
of all of our members — be it through attending or leading meetings, writing
code, or publishing benchmark scores,” said
Peter Torelli, president and CTO of EEMBC. “Each year it is our privilege to
recognize those members whose dedication and leadership has really stood out in
helping EEMBC maintain its momentum and produce compelling benchmarks with
industry support. Congratulations to our 2017 honorees and thank you for your
support.”
The honorees for 2017 are:
Ultra-Low
Power Subcommittee
·
ULPMark Working Group
·
Monica
Redon Segrara
(ADI) for taking over as group chair, resolving issues of cross-posting, and
negotiating where to steer the group next
·
Mark
Wallis and Jean-Julien
Pegoud for making the STMicroelectronics STM32 Power
Shield the official energy monitor for EEMBC benchmarks. A good deal of
co-development between EEMBC and STMicro was required
to make this happen, resulting in more sophisticated power analysis. The Power
Shield enables our ULPMark, IoTMark and SecureMark line of products.
·
Steve
Allen (Dialog Semi)
for his technical expertise, and for spearheading the effort to begin standardizing
how CoreMark® benchmark power should be collected and
reported
·
IoTMark Working Group
·
Craig
Giglio
(SiLabs) for taking over as the group chair and wrangling with the “last-mile”
challenges required to productize the benchmark
·
Steve
Allen (Dialog Semi)
for his significant efforts in porting the firmware to at least four devices,
and providing extensive testing and analysis feedback to the team. This helped
define the final profile parameters and made the code more robust
·
Mark
Wallis (STMicro) for providing technical guidance and balanced
assessment of issues facing the benchmark tuning
·
Solomon
Peachy (ARM) and Carlos
Neri Castellanos (NXP) for helping resolve the complex
debate pertaining to transmit-power certification procedures; both provided
calibrated hardware and testable hypotheses required to verify the methodology
·
SecureMark Working Group
·
Ruud
Derwig
(Synopsys) for running the working group and deftly addressing challenges to
the technical decisions made during development. He also guided the technical
content of the benchmark itself and ported it to both hardware and software,
providing further confidence in the API
·
Mike
Borza (Synopsys) for
his outreach at various conferences promoting SecureMark-TLS, and his technical
contributions to the scoring of the benchmark
·
Hannes
Tschofenig
(ARM) for his promotion of the benchmark, technical research that defined the
benchmark, conference co-presentations with EEMBC, development of the
source-code API, and debug support of the host-side validation that enabled
self-verification of the test
·
Herman
Roebbers
(Altran) for providing valuable technical input (in
all ULP workgroups), and for verifying that the SecureMark API is suitable for
hardware acceleration
Heterogeneous
Compute Subcommittee
·
ADASMark Working Group
·
Former
group chair Rafal Malewski
(formerly at NXP) for keeping the ADAS development steady and on target,
promoting the benchmark at conferences, and fielding questions from the press
·
Vlad
Calina
(NXP) for solving many significantly complex technical challenges and
consistently delivering quality software on schedule. He also provided
optimized OPenCL kernels and testing feedback during
beta
·
Cristina
Ilie
(NXP) for developing the validation strategy used by multiple stages of the
benchmark to self-check
·
Yuan
Zhao (TI) for providing technical
insight into the development of the benchmark, optimized kernels, and beta-test
support
·
Machine Learning Working Group
·
Ramesh
Jaladi,
group chair (Intel), for bringing momentum and enthusiasm to the kickoff of our
latest addition to the EEMBC benchmark family: MLMark
for machine learning on the edge
Additional
Recognition
·
Joseph
Yiu
(ARM) for his contributions in all of the ULP subcommittee working groups, and
his efforts in helping define the next-generation EEMBC performance benchmark
# # #
About EEMBC
EEMBC
develops performance benchmarks for the hardware and software used in
autonomous driving, mobile imaging, Internet of Things, mobile devices, and
many other applications. Benchmark suites are developed in a consensual process
by EEMBC member companies and EEMBC technical staff to ensure fairness of
approach and industry-wide acceptance. Further information is available at www.eembc.org.