Call for Position Papers: Extreme Heterogeneity
On behalf of the Advanced Scientific Computing Research (ASCR) program in the Department of Energy (DOE) Office of Science and ASCR Program Manager Dr. Lucy Nowell, we are organizing a Workshop on Extreme Heterogeneity (EH). Extreme heterogeneity is the result of using multiple types of processors, accelerators and memory/storage in a single computing platform that must support a variety of application workflows to meet the needs of increasingly diverse scientific domains. Extremely heterogeneous supercomputers will be acquired by the ASCR-supported computing facilities as we reach the end of Moore’s Law while still facing rapidly increasing computational and data intensive demands. The ASCR Computer Science research focus for this workshop is on system software, and software development tools and environments for supercomputers that will be delivered for operational use in the 2025-2040 timeframe.
The purpose of the EH workshop is to more clearly define the challenges that extreme heterogeneity presents to the software stack and the scientific programming environment and to identify related Computer Science priority research directions that are essential to making extremely heterogeneous systems useful, usable, efficient, and secure for science applications and DOE mission requirements.
The workshop aims to identify and prioritize research directions by analyzing existing and next-generation computer architectures. The workshop will target post-exascale architectures including novel technologies that may be developed in the "Post-Moore's Law era" and promising tools and techniques, such as advanced analytics and machine learning, that may enable efficient and productive utilization of such architectures. Participants will also discuss options to leverage methods developed by industry, such as approaches to improved developer productivity for Big Data.
We anticipate participation by personnel from universities, industry, and DOE National Laboratories. The workshop will feature a variety of plenary talks and multiple breakout sessions, with every invitee expected to participate actively in discussion of potential research directions.
Workshop participants will produce a report that will define basic needs and opportunities in computer science research to develop smart and trainable operating and runtime systems, execution models, and programming environments that will make future systems easier to tailor to the computing needs of scientists and for supercomputing facilities to securely deploy.
We invite community input in the form of two-page position papers that identify and discuss key challenges posed for supercomputing operating and runtime systems, programming models, and software development environments for scientific computing by the trend towards increasing heterogeneity in supercomputer architectures and workflows. In addition to providing an avenue for identifying workshop participants, the position papers will be used to shape the workshop agenda, identify panelists, and contribute to the workshop report. Position papers should not describe the authors’ current or planned research, nor should they recommend solutions or narrowly focused research topics. Rather, they should aim to improve the community’s shared understanding of the problem space and help to stimulate discussion.
Position papers should describe a fundamental computer science research direction that addresses key challenges associated with extreme heterogeneity.
- Key Challenges: Which challeges(s) related to Extreme Heterogeneity does this paper address?
- Research Direction: What is the promising research direction for this topic?
- State of the Art: Across the community, identify state of the art for this research direction.
This description should be followed by an assessment of potential research directions based on the following dimensions:
- Maturity: Are there existing methods or computer science research directions that address the challenge(s) and that show promise for scientific computing? What are the indicators that a given method or approach will address the identified challenges? If there are not existing methods or research approaches to meeting the challenge, can you suggest ways to gain new insight into the problem space?
- Timeliness: Why now? What breakthrough or change makes progress possible now?
- Uniqueness: Is the identified challenge unique to scientific applications of supercomputing, whether for simulation or data intensive science? What makes it so?
- Novelty: To what extent is existing method or approach unique to extreme heterogeneity in supercomputing for scientific research? Is the approach being addressed by other research programs? By the private sector? How is this approach different from existing approaches or solutions? Why should it be of interest to the ASCR Computer Science program?
Each position paper must be no more than two pages, including figures and references. The paper may include any number of authors, but must provide contact information for a single author, who could represent the position paper at the workshop. There is no limit to the number of position papers that an individual or group can submit. Authors are strongly encouraged to follow the structure outlined above. Submit position papers in PDF format at the URL listed above.
Submissions will be reviewed by the Organizing Committee. Authors of selected submissions will be invited to participate in the workshop, which will be held on January 23-25, 2018, in Gaithersburg, MD. Authors are not expected to have a history of funding by the ASCR Computer Science program. Authors of selected position papers will be invited to participate in the workshop based on the overall quality of the position paper(s) and an expectation that their active participation in the workshop will stimulate constructive discussion by the workshop participants and contribute to an informative report. Unique positions that are well presented and emphasize potentially transformative research directions will be given preference.