Fewer Faults for Faster Computing

March 28, 2011
The redundant data distribution of an array showing that a node failure will leave at least one copy of the data available for continued execution. This is the basic idea that enables the approach, which can be used in many science domains.

(PhysOrg.com) -- Environmental Molecular Sciences Laboratory (EMSL) users have designed and implemented an efficient fault-tolerant version of the coupled cluster method for high-performance computational chemistry using in-memory data redundancy.

Their method, demonstrated with the EMSL-developed and now NWChem, addresses the challenges of reduced mean time between failures, which is currently days and projected to be hours for upcoming extreme scale supercomputers.

Their approach with coupled cluster perturbative triples enables the program to correctly continue execution despite the loss of processes.

The team extended the Global Array toolkit, a library that provides an efficient and portable “shared-memory” programming interface for distributed-memory computers.

Each process in a Multiple Instruction/Multiple Data parallel program can asynchronously access logical blocks of physically distributed dense multidimensional arrays, without requiring cooperation by other processes.

The infrastructure that the team developed was shown to add an overhead of less than 10% and can be deployed to other algorithms throughout NWChem as well as other codes. Such advances in supercomputing will enhance scientific capability to address global challenges such as climate change and energy solutions using top-end computing platforms.

Explore further: Computational Science Programming Model Crosses the Petaflop Barrier

More information: van Dam HJJ, A Vishnu, and WA de Jong. 2011. “Designing a Scalable Fault Tolerance Model for High Performance Computational Chemistry: A Case Study with Coupled Cluster Perturbative Triples.” J. Chem. Theory Comput. 2011, 7, 66–75. DOI:10.1021/ct100439u

Related Stories

Scaling Goes eXtreme: Researchers reach 34K CPUs

May 25, 2010

(PhysOrg.com) -- Currently, researchers have demonstrated the scalability of high-level excited-state coupled-cluster approaches and parallel-in-time algorithms, reaching a staggering 34,000 Core Processing Units.  Researchers ...

Customizing supercomputers from the ground up

May 27, 2010

(PhysOrg.com) -- Computer scientist Adolfy Hoisie has joined the Department of Energy's Pacific Northwest National Laboratory to lead PNNL's high performance computing activities. In one such activity, Hoisie will direct ...

Recommended for you

Drone market to hit $10 billion by 2024: experts

October 3, 2015

The market for military drones is expected to almost double by 2024 to beyond $10 billion (8.9 billion euros), according to a report published Friday by specialist defence publication IHS Jane's Intelligence Review.

Radio frequency 'harvesting' tech unveiled in UK

September 30, 2015

An energy harvesting technology that its developers say will be able to turn ambient radio frequency waves into usable electricity to charge low power devices was unveiled in London on Wednesday.

Professors say US has fallen behind on offshore wind power

September 29, 2015

University of Delaware faculty from the College of Earth, Ocean, and Environment (CEOE), the College of Engineering and the Alfred Lerner School of Business and Economics say that the U.S. has fallen behind in offshore wind ...


Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.