High Performance CORBA
Working Group
Mission Statement
Mission (6/98):
The goals of the High-Performance CORBA Working Group are to identify
areas of possible standardization activities and define standards for
supporting those CORBA applications which:
are willing to specify more detailed control or hints to
achieve higher performance than that which is possible or
expected by internal ORB optimization alone (distinct
from issues of determinism or measurement and
instrumentation) - e.g. smart proxies, multi-threading,
annotations, and exposing caches;
exploit aggregated processing by using multiple processors to
accelerate a single task. Such application areas are sometimes
referred to as "cluster" or "parallel" computing. Signal
processing is an important domain of such applications. -
e.g., collective operations, distributed synchronization, and data
distribution.
Highlight (perceived or real) performance limitations of
core interfaces and recommend areas of possible change.
Additional roles may include:
Performance vetting of submissions and education with regard to
achieving higher performance in application design and implementation.
Scope:
The scope of the WG is defined in the mission statement, above. Here,
we set the High Performance CORBA WG aside other active OMG groups, by
specifically pointing out what we see not as part of the scope
of this WG.
- Real-Time SIG focusses on determinism and various aspects of QoS.
- Benchmark SIG focuses on instrumentation issues.
- Embedded WG focuses on memory footprint.
Structure:
The WG reports to the Real Time Platform SIG (RT PSIG).
Contact Information
Group Resources:
Group Chairs:
James E. Kulp
Mercury Computer Systems, Inc.
199 Riverneck Road
Chelmsford, MA 01824
US
WWW:
http://www.mc.com
E-mail: jek@mc.com
Phone: 978-256-1300
Technical Processes in Progress
Request for Information (RFI) on:
Support for Aggregated Computing in CORBA
(orbos/99-01-04. txt.
ps.
pdf
rtf,
doc.)
This RFI seeks information to help the OMG make useful and
effective decisions in the technology adoption process to
support CORBA-based applications on aggregated computing
resources, (computing clusters, multicomputers, supercomputers, network of
workstations etc.)
specifically information on issues divided into
three groups: load balancing, parallel processing, and other
standardization activities along these lines are sought.
Status: RFI has been issued; responses pending. RFI Status Page.
Past and Current Activities
Documents
RFI on Support for Aggregated Computing in CORBA
(orbos/99-01-04.
txt.
ps.
pdf
rtf,
doc.)
Presentation of RFI on Support for Aggregated Computing in CORBA
(orbos/99-01-20.
ppt,
ps,
pdf.)
Open Call for Participation
Ongoing High Performance CORBA Symposium.
Past High Performance CORBA Symposia
Burlingame, CA, US, November 9th - 13th, 1998.
Presentations:
- Message Passing Interface by Jim Kulp, Mercury;
- Performance-driven clustering and load balancing, by TBA;
- Positioning Smart Proxies against Interceptors,
by Paul Runyan, Nokia.
Seattle, US, Monday, September 14th, 1998.
Presentations:
- Fast IDL. By Victor Gidding, OIS.
- Report on: Special session on
"Middleware for High Performance Distributed
Computing", at the 1998 International Conference
on Parallel and Distributed Processing Techniques and
Applications (PDPTA'98) July 13 - 16, 1998, Las Vegas
Hilton, Las Vegas, Nevada, USA, by H.-Arno Jacobsen, HU, Berlin,
(cf. session agenda).
Orlando, US, June 1998.
General topic: High-Performance ORBs.
Presentations:
- Oisin Hurley, IONA
- Craig Lund, Mercury
Manchester, UK, March 1998.
General topic: High-Performance ORBs.
Presentations:
- Victor Gidding, OIS
- Tom Barker, Lockheed Martin
- Stefan Carrez. Sun/Chorus
TC meeting minutes
Mailing list(s):rthp@omg.org
List Archive
Miscellaneous
Following are a collection of Internet resources and mailing lists
about High-Performance Computing. If you would like to add resources,
please forward them to the chairs of the WG.
(Not yet available.)
|
