Power and Thermal Challenges in Mobile Devices
Krishna Sekar
Principal Scientist
Mobile and Wireless Group
Broadcom Corporation
Abstract:
In spite of recent advances in the development
of advanced low power designs and power management techniques, power remains
and will remain a first-class design constraint for mobile devices.
The functionality crammed into mobile devices will only continue to grow
with the increasing transition from PCs to super-phones. Power remains
one of the fundamental bottlenecks to this transition, and without
continued innovations in power management, the term "mobile" devices
will be rendered moot. In addition to (and closely related to) power challenges
in mobile devices are thermal challenges. Increasingly complex and rich
functionality in mobile devices leads to higher power dissipation and
consequently higher temperatures. However, thermal constraints are constant
across successive device generations. Thermal limits (even more than power)
will become the fundamental bottleneck to increasing the capabilities of mobile devices,
making thermal management techniques crucial. This talk discusses some
of the major challenges in power management and thermal management
for current and next-generation mobile devices from an industry perspective.
Presentation: download ( pdf)
About the Speaker:
Dr. Krishna Sekar is currently Principal Scientist
in the Mobile and Wireless group at Broadcom Corp., San Diego,
where he conducts research and development on various aspects
of system-level power and thermal issues in current and next-generation
smartphone and tablet products. Prior to this, he was with the Broadband
Communications group at Broadcom, San Diego, where he researched
and developed various embedded security solutions for Broadcom chipsets.
Dr. Sekar has served on the technical program committee of the ACM
International Conference on Mobile Computing and Networking (MobiCom)
and as an expert reviewer for numerous leading conferences and journals.
He has co-authored 10 conference and journal publications and was the recipient
of a Special Services award from the IEEE VLSI Test Symposium (2003).
Dr. Sekar received the B.Tech degree in Computer Science and Engineering
from the Indian Institute of Technology, Kharagpur, India, in 1999 and the M.S.
and Ph.D. degrees in Computer Engineering from the University of California,
San Diego, in 2001 and 2005, respectively.
| |
 |
The Alcaltel-Lucent Bell Labs Talk |
 |
 |
Towards a Virtual Cellular Network with Variable Grade Spectrum: Challenges and Opportunities
Milind M. Buddhikot
Distinguished Member of Techinical Staff (DMTS)
Alcatel-Lucent Bell Laboratories
Abstract:
This position paper presents vision
of a bold future for wireless networking research where wireless networks exploit
many variable grade spectrum bands. The transformation we envision will be fueled
by a fundamental change in the way networks use spectrum. The mix of spectrum options
will include existing exclusively licensed and unlicensed bands and new shared spectrum
bands where incumbent primary transmitters with interruptible, exclusive access share
the band with cooperating (secondary) users. Such bands used in small cell deployments
will be key to creating enormous wireless capacity. The nascent spectrum database
technologies will morph into more dynamic spectrum databases and provide essential
interference coordination, channel management and monetization. This trend when combined
with virtualization of Radio Access Network (RAN) will gradually lead to new deployment
models where entire wireless network control and data paths are collapsed into virtual
machines in a data center. Such implosion of network and democratization of spectrum
access can fuel innovative business models and new regulatory regimes for wireless networks.
We illustrate the new architecture and component radio, database and security technologies
using concrete example of incorporating shared S-band radar spectrum in a virtualized
small cell network. We also identify new business models, major research challenges
and potential roadblocks that can slow down realization of this vision into reality.
Presentation: downloload ( pdf)
About the Speaker:
Milind M. Buddhikot is currently a Distinguished Member of Technical Staff (DMTS)
in Alcatel-Lucent Bell Laboratories, where he conducts research in next generation of wireless networks.
Milind's technology contributions in areas of dynamic spectrum access, integration of Wi-Fi and cellular networks,
wireless mesh networks, and video-on-demand (VOD) services are well recognized in the research community.
Milind holds a Doctor of Science (D. Sc.) in computer science (1998) from Washington University
in St. Louis and a Master of Technology (M. Tech.) in communication engineering (1988) from I.I.T (Mumbai).
He is a recipient of the Bell Labs President's Silver Award for outstanding innovations and contributions (2003),
Bell Labs Team Award (2003), Lucent Chairman's Team Award (2006), and DMTS award (2012). He has published more than
40 technical papers in top-tier conferences and journals. He holds 9 patents and has filed over 20 other patents.
Milind is a co-founder of the IEEE DySPAN symposium which has emerged as a premier conference on the topic
of Dynamic Spectrum Access. He has served as an Associate Editor of IEEE/ACM Transactions on Networking
and Elsevier's Computer Networks Journal and regularly participates in NSF panels and TPC committees of
major IEEE and ACM conferences. More details on Milind's research can be found at
http://www.bell-labs.com/user/mbuddhikot.
Why Mobile Performance is Hard
Matt Welsh
Head of Mobile Web Performance Team
Google
Abstract:
Despite two decades of research on optimizing mobile networking performance
-- and LTE speeds pushing 50 Mbps in many markets -- the performance of mobile applications and websites
is still abysmal, especially compared to their desktop counterparts. In this talk IĄŻll explain
why this is such a hard problem, and why it has little to do with conventional research directions,
such as optimizing mobile TCP/IP performance. I'll also talk a bit about work that Google is doing in
this area on the Chrome and Android platforms.
Presentation: download ( pdf)
About the Speaker:
Matt Welsh is the head of the mobile web performance team at Google's Seattle office.
Prior to joining Google, he was a professor of Computer Science at Harvard University,
where he deployed sensor networks on volcanoes and developed programming languages
for robotic bees. These days he focuses on making the mobile web faster, cheaper, and more secure.
Designing an Energy Aware Operating System
Arun Kishan
Developmet Manager
Kernel Platform Group
Microsoft
Abstract:
TBA
Presentation: download ( pdf)
About the Speaker:
Arun Kishan joined Microsoft in 2002 and has been with the Windows Kernel team ever since.
During that time, he has worked on and led the development of various subsystems in the Windows OS and kernel,
including the process subsystem, thread scheduler, and most recently, the Windows 8 application execution model
(which in particular has been designed from the ground up with battery life in mind). He is currently the
development manager for the Kernel Platform Group, whose charter spans hardware bring up, energy efficiency,
core kernel development, and system componentization. He holds undergraduate and Masters degrees
in Computer Science from Stanford University.
|
