SenEx Statement on the Report of the Task Force on Research Computing
Approved by SenEx, February 5, 2008
The University Senate Executive Committee (SenEx) formally received the final Report of the Task Force on Research Computing at its meeting on January, 22, 2008, and discussed the work of the Task Force with the Task Force Chair, Joe Evans, and Hume Feldman, a member of the Task Force. SenEx is grateful to the Task Force and its members for the time, energy, and thought that went into its work, and for the detailed report highlighting apparent problems with the research computing situation at the University of Kansas and recommending solutions.
SenEx also has asked for comments and suggestions from the Vice Provost for Information Services as an aid to understanding the issues and how best to address them.
SenEx has some serious concerns about how the governance process worked in this case, and so cannot be certain that the Report is fully representative of the needs and concerns of the broader university community. Notwithstanding the process concerns, which will be closely examined by SenEx in other contexts, the Task Force Report stands as a strong statement of the problems and frustration felt by a number of researchers at the University of Kansas with “high end” computer needs. These concerns overlap with and mirror larger problems with the computer infrastructure at the University of Kansas, but they are also in some ways distinctive and unique to this community. Thus, while SenEx is not able to agree with every statement in the Report and is not in a position to endorse the specifics of each of the Task Force’s recommendations, SenEx agrees with and endorses the general conclusions of the Report and the principles reflected in its recommendations, which we believe are largely shared by the Vice Provost for Information Services, as follows:
1. Serious Deficiencies. The University of Kansas has serious deficiencies in its research computing infrastructure that are the cumulative result of practices and decisions over a very long period of time. These deficiencies impair the research mission of the university, including especially its ability to compete for some grants and to attract and retain researchers in various fields that involve high-end research computing needs. Three interrelated issues must be addressed to fix the problem:
2. Institutional Responsibility. Currently, there is no university-level responsibility for research computing infrastructure (including, especially, a leader who has particular expertise in this area). Under the traditional, decentralized approach, individual units, centers, and researchers fend for themselves. Over time, this approach has become increasingly ineffective and inefficient. The institutional responsibility for building infrastructure must be reflected in the organizational structure of the university through a high-level position with responsibility for developing and maintaining the research computing infrastructure for the University as a whole. While SenEx strongly endorses the Task Force recommendation for institutional responsibility to be vested in a high level official with the requisite expertise, SenEx cannot take a position on the specific recommendation of the Task Force to divide Information Services into Library and Information Technology components. That issue was not part of the Governance charges to the Task Force and has ramifications that extend far beyond the specific issues of research computing.
3. Communication. Communication between researchers with high end computing needs and Information Services is reported to be poor, with no mechanism for researchers to communicate their needs except on an individual basis. This problem relates to the institutional responsibility issue as well, since it is not clear to whom these needs should be communicated. To address the structural problems with communication, SenEx endorses the creation of an advisory body with representation from faculty and other researchers to consider and articulate research computing needs, a step which we understand to be supported by both the Task Force and the Vice Provost for Information Services. Other mechanisms to improve communication between researchers and personnel in Information Services and Research and Graduate Studies should also be considered. It is also clear that part of the problem is that researchers and administrators live in different worlds and can have trouble understanding each other. To make good communication possible, it is critical that the institutional responsibility for research computing infrastructure be led by someone with a background in the research community, as well as skill in administration.
4. Resources and Funding. At least in part because there has been no focused institutional responsibility for research computing infrastructure and poor communication between researchers and Information Services, resources are inadequate to meet high-end research computing needs. Finding the resources to develop and maintain the research computing infrastructure will be a difficult task. External funding for some aspects of infrastructure development is available, but this kind of funding must be pursued at the institutional level and requires expertise placed at that level to be successful. Funding can and must be supplemented from other sources and it is therefore essential to develop a funding model to address this issue on a long-term basis. In addressing the issue of resources and funding it is important to bear in mind that devoting resources to the larger computing infrastructure at the University of Kansas will not solve the problems of researchers with specialized high end computer needs that are central to their research agenda and that a complete solution must address both kinds of infrastructure deficiencies.
In view of the importance of these matters to the entire University community, SenEx urges the Provost and Chancellor to take the concerns expressed in the Task Force Report seriously and to address the institutional responsibilities, communication, and resource issues as quickly and comprehensively as possible. If these problems are to be addressed effectively, moreover, SenEx strongly believes it is important to for those involved, including the research community, Research and Graduate Studies, and Information Services to work together to develop and implement a unified response.