Parallel in time algorithms enable simulation of long-lasting chemical processes

Jan 27, 2014

(Phys.org) —Molecular dynamics simulations often require too much time to be practical for simulating important chemical processes that take place on long timescales. Using EMSL-developed NWChem, researchers provided evidence that time integration algorithms working in parallel can significantly speed up computationally demanding molecular dynamics simulations, opening new avenues for studying complex, long-lasting chemical processes.

New parallel in time algorithms that speed up high-level molecular dynamics simulations can enable scientists to predict the properties of complex materials. They are designed to use NWChem and other molecular simulation programs. This research is important for toxic waste disposal, nuclear waste storage, carbon sequestration, energy storage, drug discovery, disease treatment, efficient catalysis, the optimization of electronic device performance, and other applications relevant to addressing the nation's major energy and environmental problems.

Molecular dynamics simulations provide valuable information about the physical movements and interactions of atoms and molecules over time. Unlike classical approaches, ab initio (AIMD) simulations provide accurate calculations of the movements of electrons, enabling scientists to study chemical reactions that involve the breaking or formation of covalent bonds. Although AIMD simulations are useful for a wide range of important problems in areas such as liquid structure, acid–base chemistry, industrial and biological catalysis, atmospheric and aerosol chemistry, geochemistry, nanotechnology, and materials science, their use is limited because they are very computationally costly.

More time-efficient simulations have been made possible by massively parallel supercomputers and parallel algorithms, which distribute the computational workload across different processors or computers. Still, AIMD approaches currently take several months to simulate events that span picoseconds, even though many important chemical processes take much longer.

To address this problem, researchers from EMSL and the University of Chicago tested several parallel algorithms that distribute computations for different time intervals of a chemical event to different processors. These parallel-in-time algorithms sped up a conventional of 1,000 silicon atoms by a factor of three, and a challenging AIMD simulation of an atmospherically important chemical reaction involving hydrochloric acid by a factor of 14. When the AIMD simulation of this chemical process was performed on the massively parallel Chinook computer at EMSL, the Environmental Molecular Sciences Laboratory, a DOE national scientific user facility, the use of parallel in time algorithms compared with sequential algorithms reduced the duration of each computational time step from 32 seconds to seven seconds.

The parallel-in-time algorithms are suitable for cloud computing—running the programs using a large number of computers connected through a real-time communication network such as the Internet. The speedup provided by these algorithms occurred even when they were implemented on machines connected by very slow networks such as WiFi or in different time zones connected by the Internet. Moreover, these algorithms can be implemented using scripting languages such as Python as well as standard quantum chemistry packages, such as NWChem—an open-source computational chemistry code developed at EMSL. Taken together, the findings demonstrate that parallel-in-time algorithms can allow researchers to use powerful AIMD simulations to study realistic and complex that take place on long timescales.

Explore further: Learning molecular models from data

More information: Bylaska, E.J., Weare, J.Q., and Weare, J.H. "Extending molecular simulation timescales: Parallel in time integrations for high-level quantum chemistry and complex force representations." J Chem Phys. 139(7):074114 (2013). [DOI: 10.1063/1.4818328].

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 ...

Learning molecular models from data

Jan 14, 2014

Dr. Heinz Koeppl is part of a new team of scientists at IBM's Zurich research lab focused on systems biology and he is not afraid to claim that one day, soon, advanced biological processes, like cell mitosis, will be represented ...

Fewer Faults for Faster Computing

Mar 28, 2011

(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 ...

Cooling inspired by sweat

Dec 19, 2013

In many of today's electronics, the price we pay for speed comes in the form of heat. As the number of processors on a computer chip increases, so does the amount of heat each chip generates—and there's ...

Recommended for you

Scientists develop pioneering new spray-on solar cells

11 hours ago

(Phys.org) —A team of scientists at the University of Sheffield are the first to fabricate perovskite solar cells using a spray-painting process – a discovery that could help cut the cost of solar electricity.

Free pores for molecule transport

Jul 31, 2014

Metal-organic frameworks (MOFs) can take up gases similar to a sponge that soaks up liquids. Hence, these highly porous materials are suited for storing hydrogen or greenhouse gases. However, loading of many ...

User comments : 0