Workshop Themes
31 May 2015
| 4:00PM - 8:00PM | Registration & Reception |
|---|---|
| Early registration, poster set-up | |
| PI / PM interaction, light refreshment |
1 June 2015
| 8:00AM - 8:30AM | Breakfast (Falls Church) |
|---|---|
| 8:30AM - 9:00AM | Overview & Agenda (Ballston) |
| 9:00AM - 10:15AM | Project Reports (Parallel A. Cross-Cutting Approaches, Ballston) |
| 10-minute presentations | |
| 9:00AM - 9:10AM | Guang Gao |
| 9:10AM - 9:20AM | Haryadi Gunawi (Presentation) |
| 9:20AM - 9:30AM | Hyesoon Kim (Presentation) |
| 9:30AM - 9:40AM | Peter Varman (Presentation) |
| 9:40AM - 9:50AM | Sherief Reda (Presentation) |
| 9:50AM - 10:00AM | Minsui Choi (Presentation) |
| 9:00AM - 10:15AM | Project Reports (Parallel B. Foundational Principles, Falls Church) |
| 10-minute presentations | |
| 9:00AM - 9:10AM | Jonathan Appavoo (Presentation) |
| 9:10AM - 9:20AM | Joseph Devietti (Presentation) |
| 9:20AM - 9:30AM | Nathaniel Cady (Presentation) |
| 9:30AM - 9:40AM | Ryan Newton (Presentation) |
| 9:40AM - 9:50AM | Ulya Karpuzcu (Presentation) |
| 9:50AM - 10:00AM | Jose Renau (Presentation) |
| 10:15AM - 10:45AM | Break (Food & Beverage) |
| 10:45AM - 12:00AM | Panel (Applications, Ballston) |
| 3-minute position followed by open discussion | |
| Milind Kulkarni | |
| Rastislav Bodik | |
| Gul Agha | |
| Kunle Olukotun | |
| Lawrence Rauchwerger | |
|
Participants will discuss the future of parallel and
scalable computing, with a view of shaping future XPS
solicitations. To help structure the discussion, answer
five questions: (a) Applications: What are the applications that motivate future systems? What advances in algorithms and programming systems will support these applications? (b) Systems: What hardware architectures and distributed systems will support future applications? What challenges do we face in design and management? (c) Technologies: What emerging technologies will change fundamental assumptions in hardware and software? What constraints disappear? What challenges arise? (d) Methodologies: How should we perform interdisciplinary research that spans applications, systems, and technologies? How should abstraction layers evolve? (e) Risks: What are the risks that threaten the success of XPS research directions? How do we guard and hedge against these threats? | |
| 12:00PM - 1:00PM | Lunch (Falls Church) |
| 1:00PM - 2:00PM | Keynote: George Biros, University of Texas (Ballston) |
|
Algorithms and architectures for scalable N-body methods Abstract: N-body algorithms are ubiquitous in science and engineering and form the core methods for many production codes in industrial, academic, and government labs. Tree-based methods typically require irregular memory access patterns that result in reduced off- and on-node performance. Although significant progress has been made in improving off-node performance,p on-node performance remains an open problem. This is especially true for production tree-based codes that have multi-stage computations involving data reshuffles and multiple computational kernels. This on-node utilization wall -- a chronic problem since the early nineties -- not only remains unresolved but has become much more acute with the emergence of deeper memory hierarchies and manycore and heterogeneous architectures. In this talk, I will outline the computational kernels used in N-body methods and I will describe the challenges in scaling them efficiently. (Slides) | |
| 2:00PM - 3:15PM | Discussion Sessions (Parallel Sessions -- see theme assignments) |
|
Ballston -- Cross-Cutting Approaches Falls Church -- Scalable Distributed Architectures Faragut West -- Foundational Principles Vienna -- Domain-Specific Design Participants will discuss the future of parallel and scalable computing, with a view of shaping future XPS solicitations. To help structure the discussion, answer five questions: (a) Applications: What are the applications that motivate future systems? What advances in algorithms and programming systems will support these applications? (b) Systems: What hardware architectures and distributed systems will support future applications? What challenges do we face in design and management? (c) Technologies: What emerging technologies will change fundamental assumptions in hardware and software? What constraints disappear? What challenges arise? (d) Methodologies: How should we perform interdisciplinary research that spans applications, systems, and technologies? How should abstraction layers evolve? (e) Risks: What are the risks that threaten the success of XPS research directions? How do we guard and hedge against these threats? | |
| 3:15PM - 3:30PM | Break (Food & Beverage) |
| 3:30PM - 5:00PM | Discussion Sessions (Parallel Sessions -- see theme assignments) |
|
Ballston -- Cross-Cutting Approaches Falls Church -- Scalable Distributed Architectures Faragut West -- Foundational Principles Vienna -- Domain-Specific Design Participants will discuss the future of parallel and scalable computing, with a view of shaping future XPS solicitations. To help structure the discussion, answer five questions: (a) Applications: What are the applications that motivate future systems? What advances in algorithms and programming systems will support these applications? (b) Systems: What hardware architectures and distributed systems will support future applications? What challenges do we face in design and management? (c) Technologies: What emerging technologies will change fundamental assumptions in hardware and software? What constraints disappear? What challenges arise? (d) Methodologies: How should we perform interdisciplinary research that spans applications, systems, and technologies? How should abstraction layers evolve? (e) Risks: What are the risks that threaten the success of XPS research directions? How do we guard and hedge against these threats? | |
| 5:00PM - 5:30PM | First Day's Report (Ballston) |
| Session leaders report preliminary findings, recommendations for future research directions to workshop | |
| 5:30PM - 7:00PM | Reception & Poster Session (Falls Church) (Posters) |
2 June 2015
| 8:00AM - 8:30AM | Breakfast (Falls Church) |
|---|---|
| 8:30AM - 9:00AM | Overview & Agenda (Ballston) |
| 9:00AM - 10:15AM | Project Reports (Parallel A. Domain-Specific Design, Ballston) |
| 10-minute presentations | |
| 9:00AM - 9:10AM | Christopher Batten (Presentation) |
| 9:10AM - 9:20AM | Deigo Donzis (Presentation) |
| 9:20AM - 9:30AM | Matt Might |
| 9:30AM - 9:40AM | Nikos Chrisochoides (Presentation) |
| 9:40AM - 9:50AM | Qinru Qiu (Presentation) |
| 9:50AM - 10:00AM | Geoffrey Fox (Presentation) |
| 10:00AM - 10:10AM | Kamesh Madduri (Presentation) |
| 9:00AM - 10:15AM | Project Reports (Parallel B. Scalable Distributed Architectures, Falls Church) |
| 10-minute presentations | |
| 9:00AM - 9:10AM | David Wentzlaff |
| 9:10AM - 9:20AM | Richard Han (Presentation) |
| 9:20AM - 9:30AM | Shan Lu (Presentation) |
| 9:30AM - 9:40AM | Stephen Freund (Presentation) |
| 9:40AM - 9:50AM | Andy Pavlo (Presentation) |
| 9:50AM - 10:00AM | Yuqing Wu (Presentation) |
| 10:15AM - 10:45AM | Break (Food & Beverage) |
| 10:45AM - 12:00AM | Panel (Systems, Ballston) |
| 3-minute position followed by open discussion | |
| Srini Devadas | |
| Alvin Lebeck | |
| Mark Oskin | |
| Sandhya Dwarkadas | |
| Michael Swift | |
|
Participants will discuss the future of parallel and
scalable computing, with a view of shaping future XPS
solicitations. To help structure the discussion, answer
five questions: (a) Applications: What are the applications that motivate future systems? What advances in algorithms and programming systems will support these applications? (b) Systems: What hardware architectures and distributed systems will support future applications? What challenges do we face in design and management? (c) Technologies: What emerging technologies will change fundamental assumptions in hardware and software? What constraints disappear? What challenges arise? (d) Methodologies: How should we perform interdisciplinary research that spans applications, systems, and technologies? How should abstraction layers evolve? (e) Risks: What are the risks that threaten the success of XPS research directions? How do we guard and hedge against these threats? | |
| 12:00PM - 1:00PM | Lunch (Falls Church) |
| 1:00PM - 2:00PM | Keynote: David Brooks, Harvard University (Ballston) |
|
Addressing the Computing Technology-Capability Gap:
The Coming Golden Age of Design via Specialization and
Parallelism Traditional performance and energy scaling benefits based on technology improvements have slowed greatly. At the same time the demand for computing capability is unsatiated with new killer applications emerging in the domains of robotics, automotive, and machine-intelligence. Lack of progress in technology scaling will necessarily place more demands on the computer architecture and software layers to deliver capability. This talk outlines two promising directions for research in this coming Golden Age of design. First, hardware acceleration in the form of datapath and control circuitry customized to particular algorithms or applications has surfaced as a promising approach, as it delivers orders of magnitude performance and energy benefits compared to general-purpose solutions. To broaden the scope of applicability for accelerator-based architectures it will be necessary to preserve flexibility and provide programmable solutions while reducing design costs. At the same time, the last decade has seen great progress in compiler and architecture approaches to exploit parallelization. Continued research investment in this direction will lead to solutions with better scalability and even more radical approaches are on the horizon. This talk will draw highlights from several research projects in both directions. (Slides) | |
| 2:00PM - 3:15PM | Discussion Sessions (Parallel Sessions -- see theme assignments) |
|
Ballston -- Cross-Cutting Approaches Falls Church -- Scalable Distributed Architectures Faragut West -- Foundational Principles Vienna -- Domain-Specific Design Participants will discuss the future of parallel and scalable computing, with a view of shaping future XPS solicitations. To help structure the discussion, answer five questions: (a) Applications: What are the applications that motivate future systems? What advances in algorithms and programming systems will support these applications? (b) Systems: What hardware architectures and distributed systems will support future applications? What challenges do we face in design and management? (c) Technologies: What emerging technologies will change fundamental assumptions in hardware and software? What constraints disappear? What challenges arise? (d) Methodologies: How should we perform interdisciplinary research that spans applications, systems, and technologies? How should abstraction layers evolve? (e) Risks: What are the risks that threaten the success of XPS research directions? How do we guard and hedge against these threats? | |
| 3:15PM - 3:30PM | Break (Food & Beverage) |
| 3:30PM - 4:15PM | Prepare for Out-Brief (Parallel Sessions -- see theme assignments) |
|
Ballston -- Cross-Cutting Approaches Falls Church -- Scalable Distributed Architectures Faragut West -- Foundational Principles Vienna -- Domain-Specific Design | |
| 4:15PM - 5:00PM | Workshop Report (Ballston) |
| Session leaders report findings, recommendations for future research directions to workshop | |