Workshops
to be held in conjunction with
ICS'02
Workshops
Workshop and Tutorial schedule,
| |
Saturday, June 22 |
Sunday, June 23 |
| |
8:30am-12:30pm |
2:00pm-6:00pm |
8:30am-12:30pm |
2:00pm-6:00pm |
| SW Mudd 227 |
T2 |
T3 |
T4 |
T5 |
| SW Mudd 233 |
T1 |
- |
- |
- |
| SW Mudd 627 |
W11 |
W11 |
- |
- |
| SW Mudd 633 |
W22 |
W22 |
W3 |
W3 |
Notes:
- W1 will run from 9:00am - 12:30am and 2:00pm - 5:00pm
- W2 will run from 8:25am - 12:30am and 1:30pm - 6:00pm
*** BRING YOUR WIRELESS CARD TO ICS'02 ***
Thanks to Columbia University, wireless network connectivity (802.11b) will
be available in the rooms where the ICS'02 tutorials, workshops and technical
sessions will be held. We plan to have some of the tutorial and workshop proceedings
available online during the sessions, so bring your wireless card. More details
are available at http://www.columbia.edu/acis/networks/wireless/
Second Workshop on Caching, Coherence and Consistency (WC3 '02)
New York, NY, USA
to be held in conjunction with the
16th Annual ACM International Conference on Supercomputing (ICS
2002)
The workshop aims to bring together researchers
from various areas of computer science whose work is related to data caching,
coherence, and consistency. Interestingly, these three topics have been present
in the research agenda of several independent communities that usually do not
meet. The interest in these topics started in the computer architecture
community and now pervades in parallel and distributed systems research. There
have also been significant efforts to address caching, coherence, and
consistency topics using compiler, operating system, or application support. The
same topics have been addressed by the operating system community in the context
of file and storage systems for both servers and mobile systems. More recently,
the interest in these issues has been revived by the web technologies, including
content and service replication and distribution. This workshop is the first
forum to bring together people from all these areas of research by recognizing
that their specific caching, coherence, and consistency issues have common
denominators that can lead to fruitful discussions and exchange of ideas. The first
edition of this workshop, which took place in conjunction with ICS'01, was
very successful and the participants were particularly appreciative to the idea
of bringing researchers from all these areas together. This workshop continued
the tradition of the workshops on software DSM, which were associated with ICS
in 1999 and 2000.
Performance Optimization via High-Level Languages and
Libraries
The development of high-performance programs for scientific applications is
usually very complicated. The effect of the algorithm choice on memory access
costs, communication overhead etc. are often very complex. Currently available
tools for software development and performance modeling/optimization do not
provide adequate support to the developers of high-performance scientific
applications. Often, the time to develop an efficient parallel program for a
computational model is the primary limiting factor in the rate of progress of
the science. Therefore approaches to automated synthesis of high-performance
programs is very attractive and is the subject of active research at several
universities and labs now.
Breaking down the traditional separation between applications development by
domain scientists and systems software development by computer scientists, the
aim of this workshop is to bring together researchers working on several aspects
of this and related problems such as:
- program synthesis to facilitate the development of high-performance
programs for specific application domains such as signal processing,
computational chemistry, etc.
- efficient development of efficient programs from high-level mathematical
languages like MATLAB.
- development of efficient implementations of applications such as FFT for a
variety of architectures, by exploiting structural properties of the
specific application.
- automatic optimization of library implementations together with the
optimization of programs that use them.
- efficient synthesis of recursive linear algebra codes that exploit deep
memory hierarchies in current computer systems.
This workshop will be of interest of to reseachers and graduate students in
several areas such as compilation technology, domain-specific languages, library
development, problem-solving environments, etc.
We are entering a new era in computing
where we want to make it easier for users to avail of the high computing power
that is available, and for system administrators in managing the computing
resources. There is a critical need to be able to deliver systems that can
automatically detect performance bottlenecks, and dynamically adapt the
execution to fix themselves. At the same time, fault-tolerance is also an
important criterion, wherein the system automatically needs to identify any
faults and self-regulate its execution so that users and system administrators
need not be concerned with such details. A recent IBM announcement also
reiterates the importance of building such systems, which they refer to as
"autonomic computing".
This workshop is intended to bring together researchers and industrial
affiliates to begin exploring this new and challenging inter-disciplinary topic
at all levels of the system architecture within the context of high performance
computer systems.
In addition to paper presentations by researchers in this area, we are also
intending to bring in industrial speakers to give their perspective on important
research topics, and organizing a panel discussion on where future research is
really needed.
Webmaster: ics02@tc.cornell.edu