This is the 2nd August 2007 Digest of SIGARCH Messages (sigarch-aug07b):
* GPGPU 2007 Call for Papers: Workshop on General Purpose Processing Using GPUs
http://www.ece.neu.edu/GPGPU/
Submitted by Prof. David Kaeli <kaeli@xxxxxxxxxxx>
* IEEE Computer Architecture Letters: New Papers published on-line by IEEE Computer Architecture Letters
http://www.comp-arch-letters.org
Submitted by Kevin Skadron <skadron@xxxxxxxxxxxxxxx>
--Doug Burger
SIGARCH Information Director
infodir_SIGARCH@xxxxxxx
* Archive: http://lists.cs.wisc.edu/archive/sigarch-members/
* Web pages: http://www.cs.wisc.edu/~arch/www/, http://www.acm.org/sigarch/
* To remove yourself from the SIGARCH mailing list:
mail listserv@xxxxxxx with message body: unsubscribe SIGARCH-MEMBERS
-----------------------------------------------------------------
Doug Burger Office: 3.432 ACES
Associate Professor Phone: 512-471-9795
Department of Computer Sciences Assistant: 512-232-7460
The University of Texas at Austin Fax: 512-232-1413
1 University Station, #C0500 E-mail: dburger@xxxxxxxxxxxxx
Austin, TX 78712-1188 USA www.cs.utexas.edu/users/dburger
----------------------------------------------------------------------
----------------------------------------------------------------------
* GPGPU 2007 Call for Papers: Workshop on General Purpose Processing Using GPUs
******************************
CALL FOR PAPERS
******************************
Workshop on General Purpose Processing Using GPUs
Northeastern University
Boston, MA USA
October 4, 2007
Overview: The goal of this workshop is to provide a forum these
general-purpose purpose programming environments and platforms,
as well as discuss applications that have been able to harness
the horsepower provided by these platforms. This year's work is
particularly interested in imaging applications. Papers are being
sought on many aspects of GPUs, including (but not limited to):
+ GPU applications + GPU software and operating systems
+ GPU programming environments + GPU power/efficiency
+ GPU architectures + GPU benchmarking/measurements
Paper Submission: Authors should submit a 8 page paper in
IEEE double-column style to gpgpu@xxxxxxxxxxxx
Industry Participation: The workshop encourages participation by
GPU manufacturers, software vendors, or companies which develop or market
products used by the GPU community. Any company interested in participating
in the workshop should contact the workshop organizer at gpgpu@xxxxxxxxxxxx
Important Dates:
Paper submission: August 28, 2007
Author notification: September 7, 2007
Final paper: September 14, 2007
Copies of final papers will be made available at the workshop.
In addition, selected papers will be invited to be part of a special
issue of an ACM or IEEE journal or magazine.
For more information, see the GPGPU 2007 web
page: http://www.ece.neu.edu/GPGPU/
----------------------------------------------------------------------
----------------------------------------------------------------------
New papers published online by IEEE Computer Architecture Letters
Computer Architecture Letters announces our three most recent papers,
which are available now via IEEE Xplore. We happy to report that, as recognition of
Computer Architecture Letters has grown, the submission rate has nearly doubled this year.
We continue to seek new submissions and remain committed to fast and accurate review. Our
mean time to decision in 2007 has been 24 days, with an acceptance rate of approximately 25%.
For more information on submission, please see http://www.comp-arch-letters.org
- J. Zebchuk and A. Moshovos. "A Building Block for Coarse-Grain Optimizations in the
On-Chip Memory Hierarchy." Computer Architecture Letters, vol. 6, June 2007.
- A. Das, S. Ozdemir, G. Memik, J. Zambreno, and A. Choudhary. "Microarchitectures for
Managing Chip Revenues under Process Variations." Computer Architecture Letters, vol. 6,
June 2007.
- J. Kim, J. Balfour, and W. Dally. "Flattened Butterfly Topology for On-Chip Networks."
Computer Architecture Letters, vol. 6, July 2007.
Abstracts
---------
- J. Zebchuk and A. Moshovos. "A Building Block for Coarse-Grain Optimizations in the
On-Chip Memory Hierarchy." Computer Architecture Letters, vol. 6, June 2007.
Abstract:
Current on-chip block-centric memory hierarchies exploitaccess patterns at the fine-grain
scale of small blocks.Several recently proposed memory hierarchy enhancementsfor coherence
traffic reduction and prefetching suggest thatadditional useful patterns emerge with a
macroscopic,coarse-grain view. This paper presents RegionTracker, adual-grain, on-chip
cache design that exposes coarse-grainbehavior while maintaining block-level communication.
RegionTracker eliminates the extraneous, often imprecisecoarse-grain tracking structures of
previous proposals. Itcan be used as the building block for coarse-grainoptimizations,
reducing their overall cost and easing theiradoption. Using full-system simulation of a
quad-core chipmultiprocessor and commercial workloads, we demonstratethat RegionTracker
overcomes the inefficiencies of previouscoarse-grain cache designs. We also demonstrate
how RegionTracker boosts the benefits and reduces the cost of apreviously proposed snoop
reduction technique.
- A. Das, S. Ozdemir, G. Memik, J. Zambreno, and A. Choudhary. "Microarchitectures for
Managing Chip Revenues under Process Variations." Computer Architecture Letters, vol. 6,
June 2007.
Abstract:
As transistor feature sizes continue to shrink intothe sub-90nm range and beyond, the
effects of process variationson critical path delay and chip yields have amplified. A
commonconcept to remedy the effects of variation is speed-binning, bywhich chips from a
single batch are rated by a discrete range offrequencies and sold at different prices. In
this paper, we discussstrategies to modify the number of chips in different bins andhence
enhance the profits obtained from them. Particularly, wepropose a scheme that introduces
a small Substitute Cacheassociated with each cache way to replicate the data elementsthat
will be stored in the high latency lines. Assuming a fixedpricing model, this method
increases the revenue by as much as 13.8% without any impact on the performance of the chips.
- J. Kim, J. Balfour, and W. Dally. "Flattened Butterfly Topology for On-Chip Networks."
Computer Architecture Letters, vol. 6, July 2007.
Abstract:
With the trend towards increasing number of cores in a multicore processors, the on-chip
network that connects the cores needs to scale efficiently. In this work, we propose the
use of high-radix networks in on-chip networks and describe how the flattened butterfly
topology can be mapped to on-chip networks. By using high-radix routers to reduce the
diameter of the network, the flattened butterfly offers lower latency and energy consumption
than conventional on-chip topologies. In addition, by properly using bypass channels in the
flattened butterfly network, non-minimal routing can be employed without increasing latency
or the energy consumption.
|