9/17/12: X-Stack Portfolio Kickoff Meeting

Agenda

The Agenda is here.

The meeting includes principal investigators selected for funding in the 2012 X-Stack Funding Opportunity as well as key contributors from synergistic Exascale research efforts. Additional information and guidance about this meeting will be provided soon.

Significant contributors of each project will:

1. Share vision and technical details of their projects;
2. Discuss collaboration and leveraging of research deliverables of the X-Stack portfolio;
3. Help create a complete vision for X-Stack software research;
4. Help develop alternatives for the X-Stack architecture and interfaces.

We expect that these activities will be extended to include Exascale research efforts carried by participants who are not X-Stack awardees.

Presentations

• X-­‐Stack Kickoff Meeting - Sonia R. Sachs
• Brandywine: Innovations in Programming Models, Compilers, and Runtime Systems for Dynamic Adaptive Event-Driven Execution Models
• Overview of CESAR Proxy Apps
• Exascale Co-design Consortium (ECDC): Operations Plan
• CORVETTE: Program Correctness, Verification, and Testing for Exascale
• DEGAS: Dynamic Exascale Global Address Space
• DSL Technology for Exascale Computing (D-TEC)
• A Brief Introduction to the ExaCT Co-Design Center
• Exploiting Global View for Resilience (GVR) - An Outside-­‐‑In Approach to Resilience
• SLEEC: Semantics-rich Libraries for Effective Exascale Computation
• Traleika Glacier (X-Stack)
• eXascale Programming Environment and System Software (XPRESS)
• Exascale Co-Design Center for Materials in Extreme Environments (ExMatEx)
• X-Stack Front-End
• X-TUNE - Autotuning for Exascale: Self-Tuning Software to Manage Heterogeneity

Project Handouts

Include in your project handout the URL of your project website. The requirement for your project website was communicated in your award letter. The project website should include sufficient information about your project to enable the community to clearly understand what will be accomplished, and it should include contact information for the different components of your project.

• Brandywine: Innovations in Programming Models, Compilers and Runtime Systems for Dynamic Adaptive Event-­‐Driven Execution Models
• CORVETTE: Program Correctness, Verification, and Testing for Exascale
• DEGAS: Dynamic Exascale Global Address Space
• D-TEC – DSL Technology for Exascale Computing (1-pager)
• D-TEC – DSL Technology for Exascale Computing (2-pager)
• GVR - Exploiting Global View for Resilience
• SLEEC: Semantics-rich Libraries for Effective Exascale Communication
• Traleika Glacier (X-Stack) Project
• XPRESS: eXascale Programming Environment and System Software
• X-TUNE - Autotuning for Exascale: Self-Tuning Software to Manage Heterogeneity

Project Posters

• Brandywine: Innovations in Programming Models, Compilers, and Runtime Systems for Dynamic Adaptive Event-Driven Execution Models
• Corvette: Program Correctness, Verification and Testing for Exascale
• DEGAS: Dynamic Exascale Global Address Space
• DTEC: DSL Technology for Exascale Computing
• GVR: Exploiting Global View for Resilience
• SLEEC: Semantics-rich Libraries for Effective Exascale Computation
• Traleika Glacier (X-Stack) – Intel Team
• XPRESS: eXascale Programming Environment and System Software

Sessions

The parallel and panel sessions for our kickoff meeting depend on the questions that we, as a group, decide we need to address during the time allocated for these sessions. We will decide on 5-6 questions for the parallel sessions, and on 4 questions for each panel session.

For the 3 panel sessions, there are at least five panelists, each presenting four slides, one for each selected panel question. The envisioned process is the following:

• The moderator asks three panelists to answer question 1 (about 2 minutes per panelist, 6 minutes total)
• A 9 minute discussion among panelists, with questions from the audience follows.
• The moderator repeats the process for question 2 – 4.

The parallel sessions are working sessions: all participants should plan to contribute to the discussions. A moderator ensures that no one dominates the discussion and that all have the opportunity to speak and contribute.

Panel Session 1

Panel Session 1: Dan Quinlan Roles: Milind Kulkarni, David Padua, Andrew Lumsdaine, John Shalf, Vivek Sarkar, John Bell; and Tina has agreed to be our scribe.

Questions:

1. What is the role of DSLs within Exascale and what impact does this have on vendors?
2. Will we see language extensions or new parallel languages from the Exascale program?
3. Where in the software stack is resilience best addressed; or need it be addressed uniformly everywhere?
4. What is the role of correctness in the development of Exascale software?
5. How fast will hardware change, how fast do we expect software and programming models to adapt? (the software stack?)

Panel Session 2: Runtime Systems

Panel Session 2: Saman Amarasinghe Panelists: Andrew Chien, Vijay Saraswat, Hartmut Kaiser, Vivek Sarkar, Wilf Pinfold

Questions:

1. We first need a common vocabulary to discuss the complex runtime system, which has many components. Can you come-up with a clear set of components for a runtime system (akin to the ISO 7 layers for networking)
2. Can you explain your runtime solution w.r.t. these components
3. In your opinion, what components are closes to get standardized (i.e. shared across all groups) and what needs lot more research/exploration?
4. What are the top three challenges in building a runtime system exascale?
5. Share some of the key ideas/technology you think will address these challenges.

Panel Session 3: Low-Level Representation: Synthesis, Refinements, Transformations, Resilience

Moderator: Armando Solar-Lezama (MIT) Scribe: Tina Macaluso (SAIC) Panelists: Koushik Sen (UC Berkeley), David Padua (UIUC), Mary Hall (Utah), Milind Kulkami (Purdue), Richard Lethin (Reservoir Labs), Vijay Saraswat (IBM)

Questions:

1. How can we get the programmer involved in refinement?
2. How do we introduce automation without losing control?
3. How do you provide situational awareness to the programmer guiding the refinement?
4. What features are needed in the program representation to facilitate synthesis, transformation and refinement?
5. How do we introduce learning into the refinement process? Should we?

Parallel Session I: Kathy Yelick

Roles: Kathy Yelick will moderate, Tina Macaluso will scribe, Mary Hall will report out.

Presentation: X-Stack Frontend

Questions:

This session will be split into two discussions, the first on the interactions with application scientists (aka X-Stack users) and the second on the languages, compilers and tools. Here are some questions:

Part 1:

1. How should we (X-stack) interact with users? This is both a technical question of what kinds of APIs / programming models they will accept and how we should interact with them in the mean time.
2. Are different X-Stack projects addressing different types of users?

Part 2:

1. How should memory and network hierarchy be managed?
2. How can we predict the importance of locality at different levels, and which things should be exposed to users? Can we "virtualize" the hierarchy using a recursive model?
3. Can DSLs help substantially?
4. Should we be writing programs that generate code rather than code itself, i.e., train more compiler folks or build tools that makes this easier.
5. Are there different types of tools that should be provided for exascale users?

One of the goals of this breakout is to collect a matrix of the approaches being used by each project.