Volume 7, Issue 1 -
Spring/Summer 1999
Volume 6, Issue 3
Fall 1998
Volume 6, Issue 2
Spring/Summer 1998
Volume 6, Issue 1
Winter 1998
Volume
5, Issue 4
Fall 1997
Volume
5, Issue 3
Summer 1997
Volume
5, Issue 2
Spring 1997
Volume
5, Issue 1
Winter 1997
Volume
4, Issue 4
Fall 1996
Volume
4, Issue 3
Summer 1996
Volume
4, Issue 2
Spring 1996
Volume
4, Issue 1
Winter 1996
Volume
3, Issue 4
Fall 1995
Volume
3, Issue 3
Summer 1995
Volume
3, Issue 2
Spring 1995
Volume
3, Issue 1
January 1995
Volume
2, Issue 4
October 1994
Volume
2, Issue 3
July 1994
Volume
2, Issue 2
April 1994
Volume
2, Issue 1
January 1994
Volume 1, Issue 4
October 1993
Volume
1, Issue 3
July 1993
Volume
1, Issue 2
April 1993
Volume
1, Issue 1
January 1993
|
High-Performance Computing in Trouble
Ken Kennedy, Director, CRPC
This year's appropriations debate in Congress has highlighted the extent
to which the High Performance Computing and Communications (HPCC)
program has fallen from grace. In the Senate, the appropriation for the
National Science Foundation (NSF) HPCC program was explicitly reduced by
$50 million, "out of concern over the Foundation's inability to
articulate specific, quantifiable, and measurable goals for its
activities in high-performance computing." On the House side, the HPCC
budget for the Defense Department's Advanced Research Projects Agency
(ARPA) was marked for a $100 million reduction from the President's
request. Although financial limitations were cited as the reason for the
ARPA cut, language accompanying the bill explicitly mentioned concern
raised by reports from the General Accounting (GAO) Office and the
Congressional Budget Office (CBO). In addition, the House appropriations
report asserted that the ARPA program was spending too much money on
acquisition of hardware for which the software was not yet mature. As I
am writing this editorial, the conference committee for the NSF has
voted for a smaller $12.5 million HPCC reduction, and the Senate version
of the appropriation for ARPA included a $53 million dollar reduction
for the Scalable Computing and Microsystems projects. While these
smaller reductions would be less destructive than those originally
proposed, they will nevertheless severely impede progress in the HPCC
program.
It is worth considering how HPCC could get into so much trouble only two
years after it was endorsed by a near-unanimous vote in both houses of
Congress. It is particularly puzzling because Al Gore, the author of the
original bill, has become the vice president and is in charge of the
administration's technology program. I believe that a large part of the
problem is the poor quality of information that Congress is being given
about high-performance computing. The Congressional Budget Office report
on HPCC, which I criticized in the July issue of Parallel Computing
Research, is a good example. This report was written by an economist who
failed to consult a single recognized expert on high-performance
computing technology in his research. Furthermore, he ignored critical
feedback from the FCCSET committee on early drafts of the report. As a
result, he mistakenly concluded that funding for high-performance
computing research was benefiting only companies that produced high-end
parallel supercomputers and hence, was not worth the cost. The irony of
this situation is that the ARPA program most affected by the reductions
spends more than 50 percent of its budget on software research.
Therefore, the cuts are likely to impede progress on software for high-
performance systems-- exactly the opposite of what Congress hopes to
achieve.
I firmly believe that the Congress and its staff are, on the whole, well
-meaning and intelligent people. Unfortunately, bad information leads to
bad decisions, no matter how intelligent the decision-maker. As a
community, we need to do a much better job of getting our story across
to Congress. The story is a good one--the high-performance computing
research community is addressing important problems and actively
transferring the results of their research to industry and to scientists
and engineers who use high-performance systems. Given the concern in
Washington with job creation, it will not be enough for researchers to
convey the message directly to Congress. To have any real impact, it
must come from corporations that are the beneficiaries of HPCC research.
Table of Contents
|