Workshop on Parallelization of Coupled-Cluster Methods

Sponsored by NSF and DOE

Saturday February 23, 2008 2-7 pm

 

2:00-2:40 Opening lecture                             Ryan Olson, Cray Research

 

2:40-2:50 Questions

 

2:50-3:30 Panel 1 Opening statements         Parallel efficiency

Topic:

Impact of communication protocols and memory use, MPI-1, MPI-2 and one-sided communication, shared memory and OpenMP, input-output to disk storage from parallel programs.

Panel members:

            Curt Janssen

            Tomasz Janowski

Wibe de Jong

Thom Dunning

3:30-4:00 Panel 1 Discussion                        Parallel efficiency

 

4:00-4:20 break

 

4:20-5:00 Panel 2 Opening statements         Parallel Coupled Cluster

Topic:

Issues related to parallel implementation of algorithms to solve coupled-cluster equations, what is similar to and different from algorithms used in other large-scale parallel software, e.g. molecular dynamics, finite element analysis, weather simulation.

Panel members:

Josef Noga

Karol Kowalski

John Watts

5:00-5:30 Panel 2 Discussion                        Parallel Coupled Cluster

 

5:30-5:50 Break

 

5:50-6:30 Panel 3 Opening statements         Benchmarks

Topic:

Can we define a set of problems that everyone can run with their software on their hardware and report to the community? Similar to the Linpack benchmark in linear algebra, we need some molecules and some properties, energy, gradient, with a list of methods. Maybe also a molecule like a polymer that can be       made larger in a defined way to provide more work             as the number of processors is increased

Panel members:

            Peter Pulay

            Sudhakar Parmidighantam

P. (Saday) Sadayappan

6:30-7:00 Panel 4 Discussion                        Benchmarks

 

Sunday February 24, 2008 1-5 pm

 

1:00-1:40 Panel 4 Opening statements         Petascale algorithms

Topic:

What algorithms can we use to scale CC methods to 1,000, 10,000 and 100,000 processors? Do we need all new algorithms? How do we measure performance? How do we debug?

Panel members:        

Ed Valeev

So Hirata

            Shawn Brown

Charles Patterson

Beverly Sanders

1:40-2:10 Panel 4 Discussion                        Petascale algorithms

 

2:10-2:30 Break

 

2:30-5:00 Conclusions and recommendations

Discussion of the panel topics and formulation of conclusions and recommendations of the workshop. One of the goals of the workshop is to present a report that defines the present status and future needs and direction of parallelization of Coupled-Cluster methods.