Title: HPC University BoF (Birds-of-a-Feather Session)
1HPC University BoF(Birds-of-a-Feather Session)
- Presenting the Preliminary Analysis of
- High Performance Computing
- Training and Education
- Requirements and Recommendations
2Session Agenda
- HPC University Introduction
- HPC Overview
- HPC University RAT Report
- Breakout Discussions
- Summary
- Next Steps
3Introduction What are we working on?
- HPC University RAT Charter
- The HPC Training RAT will identify successful
paths to creating qualified, effective HPC
professionals, capable of exploiting current
super-terascale and upcoming petascale
technologies for the advancement of scientific
research. Gaps in training materials and delivery
methods will be identified, and recommendations
for filling these gaps will be presented. - RAT Requirements Analysis Team, an ad hoc
group of TG staff and others, with a short-term
mandate, who are chartered to evaluate a specific
topic and report to TG management on feasible
solutions. NOT an implementation team.
4Whos Involved?
Laura McGinnis Chair Pittsburgh Supercomputing
Center
Members Bonnie Bracey-Sutton Diglio
Simoni (GLEF) (RTI International) John
Cobb Stephenie McLean (ORNL) (RENCI) Shawn
Brown Christian Chilan (PSC) (UC/ANL)) Brad
Armosky Sudhakar Pamidighatam (UT
Austin/TACC) (NCSA) James Rome John
Ziebarth (ORNL) (Krell Institute) Mary Ann
Leung Scott Lathrop (e-SciTek, LLC) (UC/ANL) Kay
Hunt Subha Sivagnanam (Purdue) (SDSC) Krishna
Madhavan (Clemson)
Advisory Alina Bejan (Open Science Grid) Steve
Gordon (Ohio Supercomputing Center) Leslie
Southern (Ohio Supercomputing Center) Robert
Panoff (Shodor) Kristina Wanous (UC/ANL)
5Why Are We Doing This?(HPC Overview)
6Terascale ? PetascaleAre you ready?
7HPC Resources Supercomputer Capability Growth
Gr Challenges in Science Engineering
HumanGenome2003
10 P
1,000 T
1,000 T
500 T
2 Years 180 T-FLOPS
100 T
4 Years 90 T-FLOPS
10 T
4 Years 9 T-FLOPS
1 T
4 Years 0.9 T-FLOPS
0.1 T
4 Years 0.09 T-FLOPS
0.01 T
5 Years 0.009 T-FLOPS
0.001 T
100 M
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
1985
1984
2015
Cray X-MP4 processors
8Earthquake SimulationGrand Challenge
Remote Runtime Steering of Integrated Terascale
Simulation Visualization
Hongfeng Yu, Tiankai Tu, Jacobo Bielak,Omar
Ghattas, Julio C. Lopez, Kwan-Liu Ma,David R.
OHallaron, Leonardo Ramirez-Guzman,Nathan
Stone, Ricardo Taborda-Rios, John Urbanic
9Terascale?PetascaleAre you ready?
10HPC University Vision
- Prepare next generation of users for next
generation of computers - Training road map/ pathways
11Training road map
12Mapping core competencies
13Methodologies
14What Have We Learned (so far)?(HPC University
RAT Report)
15RAT Report
- Catalog Map
- Topics
- Mastery Levels
- Quality Assurance
- Delivery Methodologies
- Target Populations
- Demographics
- Disciplines
- Scaling Dissemination
16General Topics
Modeling Simulation HPC Technology
(hardware) Architectures (Parallel, Dist,
Grid) Workflow management Programming
Algorithms Development Tools Software
Engineering Operational Issues Performance
Analysis Verification Validation Code
Optimization Science gateways resources Data
analysis/post processing Visualization Scalable
Computing Domains (physics, chem, etc.) HPC
Application packages Collaboration
At least 10 Less than 10 None
17Catalog Map
At least 10 Less than 10 None
18Catalog Map Recommendations
- Complete the verification and validation process
for all entries in the catalog - Develop a mechanism for the continual update of
the catalog to identify new sources of training
materials - Make the catalog readily available to the HPC
community
19Quality Assurance
- CSERD/NSDL Overview
- Digital repository for instructional reference
materials for computational science resources - Includes comprehensive review mechanism for
materials, similar to journal review process - Solving the problem correctly ! Solving the
correct problem? - Verification
- the resource works as advertised on the computing
platforms as advertised - Validation
- the resource is based on current, valid
scientific methods - Accreditation
- the resource is appropriate for the advertised
audience
20Quality Assurance Recommendations
- Leverage CSERD/NSDL
- Broad dissemination
- Formal VVA review process.
- Resource Management i.e. identify a primary
editor - Finding reviewers,
- Organizing reviews
- Publishing the resources that have passed a
minimum quality level - Broadening Community Involvement
- Contribute additional materials
- Offer community comments on the resources
- Provide feedback on the resources to TeraGrid on
a regular basis
21Delivery Methodologies
Total number of offerings 239
This is a hole
22Delivery Methodologies Recommendations
- Plug the holes - provide a variety of teaching
methods for each topic area - Allow students to customize training programs to
meet their learning styles. - Redesign course materials to produce effective
interactive online learning experiences - Avoid simply putting lecture-based materials
online - Utilize efficient multi-disciplinary teams,
including - subject matter experts
- instructional designers technologists
- information technologists
- tutors
- faculty
- project managers
- Incorporate metacognitive skills development
into all training methods - Create or adopt a module that teaches students
how to become aware of their most effective
learning styles - Supplement these materials with guidelines on how
to select the appropriate training materials
based on their learning styles
Metacognitive the act of thinking about
thinking
23Population Analysis - Demographics
- Underrepresented populations The
underrepresented populations are measured as the
number of users at MSIs
- The MSI list is generated from a number of
sources - ed.gov (the US Department of Education) list of
MSIs - collegeboard.com (college search site), with
search criteria of - 4-year HBCUs
- 4-year Hispanic schools
- 4-year Womens schools
- NSF EPSCOR list
Other
Institutions (499 total)
Users (6475 total)
EPSCOR
Hispanic
HBCU
Womens Colleges
24Population Analysis - Disciplines
- Represented Fields of Science
- Molecular Biosciences Materials Research
- Astronomical Sciences Advanced Scientific
Computing - Mechanical Structural Systems Earth Sciences
- Integrative Biology Neuroscience Electrical
Communication Systems - Chemistry Chemical, Thermal Systems
- Physics Atmospheric Sciences
- Computer Computation Research Mathematical
Sciences - Biological Critical Systems This represents 90
of all projects allocated to the TeraGrid.
- Underrepresented Fields of Service
- Design Manufacturing Systems Training
Inorganic, Bioinorganic, Organometallic
Chemistry - Biophysics Networking Communications
Research Theoretical Physics - Behavioral Neural Sciences Algorithm
Development Metals, Ceramics, Electronic
Materials - Performance Evaluation Benchmarking Ocean
Sciences Gravitational Physics - Analytical Surface Chemistry Elementary
Particle Physics Extragalactic Astronomy
Cosmology - Nuclear Physics Software Systems Statistics
Probability - Thermal Systems Distributed Parallel
Processing, Vectorization Educational
Infrastructure - Genetics Nucleic Acids Global Atmospheric
Research Environmental Biology - Neuroscience Biology Planetary Astronomy Solar
Terrestrial Research - Social Economic Science Cross-Disciplinary
Activities Information, Robotics, Intelligent
Systems - Physical Chemistry Biochemistry Molecular
Structure Function Galactic Astronomy - Fluid, Particulate, Hydraulic Systems Applied
Mathematics Organic Macromolecular Chemistry - Condensed Matter Physics Engineering
Centers Science Engineering Education - Atmospheric Chemistry Engineering Infrastructure
Development Mechanics Materials - Polar Programs Solid State Chemistry
Polymers Stellar Astronomy Astrophysics - Biological Instrumentation Resources
Economics Microelectronic Information
Processing Systems - Geophysics Industrial Science Technological
Innovation Meteorology - Physiology Behavior Quantum Electronics, Waves,
Beams Solid-State Microstructures
25Population Analysis Recommendations
- Proactively support efforts to Broaden
Participation across - Underrepresented demographics
- Underrepresented fields of science
- Broaden delivery methods
- Synchronous eliminates geographic boundaries
- Asynchronous eliminates temporal boundaries
26Scaling Dissemination
- Train-the-trainer
- Most people teach the way they were taught
- BUT, most survey respondents are self-taught
- Sohow do we scale and disseminate HPC training?
- Collaboration Mentoring
- Time-intensive for master, but effective
methodology (long history of success) - Traditionally geographically temporally
constrained - Technology-du-jour - Masters may be disinclined
to learn new technology - Learning curves
- Versioning dialects
- Pain threshold for dealing with reliability
issues
27Scaling Dissemination Recommendations
- Identify and promote Best Practices for content
development and delivery - Encourage mentoring within TeraGrid
- Create a Web portal for people to volunteer their
services - Create a Web Portal for users to ask questions
and to get a mentor - Provide real incentives for such volunteer
activity - Study the social interactions in mentoring
situations to provide training to make mentoring
more effective - Provide support to HPC trainers
- Training accounts on TeraGrid
- Help with developing and critiquing course
content - Provide a Web Portal for users new to HPC that
would provide a roadmap of training that would
lead them to their individual goals
28How can YOU help?(Breakout Sessions)
29Breakout Groups
- Catalog
- Quality Assurance
- Delivery Methodologies
- Populations
- Scaling Dissemination
30Breakout Groups Questions to Answer
- Are these the right topics?
- What are we missing?
- Are the recommendations sensible? Feasible?
- What would you like the next steps to be?
31Where do we go from here?(Next Steps)
32RAT Next Steps
- Consolidate info from SC07
- Present to TG governance (December 2007)
- On approval, begin implementation
- Project plan
- Funding
- Rollout
- Etc