|Volume 7, Issue 1 -
A DEFINITIVE REPORT ON HPCC
Ken Kennedy, Director, CRPC
Over the past two years, there have been so many studies and reports on the High Performance Computing and Communications Initiative (HPCCI) that one wonders how anyone has time to do the research that is the main goal of the program. Recently, however, a report has been published that was well worth the time and effort. It merits widespread public attention, support, and response.
The report, entitled "Evolving the High Performance Computing and Communications Initiative to Support the Nation's Information Infrastructure," comes as a valuable clarification at a time when public policy regarding HPCCI is confused by many differing opinions about its effectiveness and direction. It comes at a time when many suspect that however laudable the original goals, somehow the HPCCI is not working.
There are two key reasons why a concerned public -- including public policymakers seeking clear, coherent ways to manage the program's future -- should pay attention to this report over the many others that have appeared. First, it is the product of a Congressionally-mandated effort by the Computer Science and Telecommunications Board (CSTB) of the National Research Council, an arm of the National Academies of Science and Engineering known for its authoritative, scholarly approach to important policy issues. Second, representing impeccable credentials and a broad spectrum of backgrounds, the committee that wrote the report carried out a thorough investigation of the program and interviewed people from all backgrounds and points of view. They even interviewed Phil Webre, author of the infamous Congressional Budget Office Report on the HPCCI, a document that might have been more accurate had Mr. Webre taken the time to interview anyone of a stature comparable to the authors of the CSTB report.
The report states that as industrial labs downsize, it is critical for the future economic health of the nation that university-based researchers collaborate with industry. The report validates my assertion that policymakers in this era of government downsizing must understand that computer science research is essential to the continued economic health of the nation. HPCC is one area where the United States leads the world and we must not squander that lead.
In addition, the report correctly concludes that the HPCCI has had many important accomplishments, but the time has come to shift its emphasis toward research that will more directly impact the National Information Infrastructure, while also maintaining a strong program in high- performance computing. Effectively addressing both research areas will stimulate work that will lead to better access to information and services of the NII, while assuring that there will be sufficient computational cycles to provide those services.
One other thing the report calls for is retaining the beneficial distribution of the HPCCI management across several federal funding agencies. However, it also recommends that the National Coordinating Office be strengthened to ensure strong collaborations among various agency programs. Implementing this strategy should result in an increase in the number of projects that are jointly sponsored by several agencies. Examples of such projects today include the National HPCC Software Exchange and the Scalable I/O consortium. CRPC researchers provide leadership for both of these projects.
The report's message is clear and constructive -- the HPCC initiative has generally been a success, but a course correction is now needed to ensure that the program continues to meet the needs of the nation in the light of changing technology. While everyone may not agree with each specific recommendation this report makes, its statesmanlike commentary on the impact and directions of the HPCC program is a welcome rallying point for keeping our country's preeminence in high-performance computing and communications.
Table of Contents