Harold Varmus Requests Input on NCI Cloud Initiative
Letter to All NCI Grantees, April 2013
To the Cancer Research Community:
The advent of a variety of new research technologies has dramatically increased the rate at which biological data is generated, a rate of change that has strained conventional mechanisms for distributing and analyzing this information. The NCI (and the biomedical community broadly) is investigating next generation computational capabilities to support the biomedical research community. One possibility is the creation of public "cancer knowledge clouds", namely data repositories with co-located computational resources, allowing investigators to bring their analytic tools to the data rather than move the data to their analytic tools. Such clouds have the potential to increase the speed of discovery and democratize access to cancer genomics data, which is too often the province of organizations that can support the high cost of maintaining these enormous data sets. While there is an emerging consensus that such computational resources are going to be essential components of the resource environment in the near future, there is not yet a consensus on the implementation of such an environment. NCI believes that the correct approach is to conduct a series of pilots that will lead to the development of a cancer knowledge cloud or clouds.
In order to ensure that our initial efforts in this area are directed at the practical problems faced by biomedical researchers, we are writing to request your advice about the scientific questions that should guide the process of developing infrastructure to analyze these high volume data sets. Specifically, we are requesting that you provide us with information on:
- Situations where limitations in information technology capabilities are either preventing you from carrying out certain types of high value research, or where these limitations, while not preventing your research, are slowing the pace of discovery significantly.
- Your experiences using high performance computing environments for biological research, including custom environments and commercial clouds (private or public).
- Metrics that could be utilized to determine the level of success of pilot "cancer knowledge clouds".
Your responses (along with those from other scientists and relevant oversight committees) will help to guide the NCI in its efforts to ensure that the research community has the tools it needs. If possible, we would like to receive your comments by May 3, 2013; they can be sent electronically to firstname.lastname@example.org.
Thank you for your prompt attention to this matter.
Harold E. Varmus, M.D.
George A. Komatsoulis, Ph.D.
Center for Biomedical Informatics and Information Technology
Louis M. Staudt, M.D., Ph.D.
Center for Cancer Genomics
Stephen J. Chanock, M.D.
Center for Cancer Genomics