Efficiency, parallel

Esselink K, Smit B and Hilbers P A J 1993 Efficient parallel implementation of molecular dynamics on a toroidal network. I. Parallelizing strategy J. Comput. Phys. 106 101-7... 

R. Murty and D. Okunbor, Efficient parallel algorithms for molecular dynamics simulations , submitted to Parallel Computing. 

NWChem (we tested Version 3.2.1) is a program for ah initio, band-structure, molecular mechanics, and molecular dynamics calculations. The DFT band-structure capability is still under development and was not included in the Linux version tested. NWChem is unique in that it was designed from scratch for efficient parallel execution. The user agreement is more restrictive than most, apparently because the code is still under active development. At the time of this book s publication, limited support was available for users outside of the EMSL facility. 

Proper load balance is a major consideration for efficient parallel computation. Consider a job distributed over two processors (0 and 1) in such a way that wall clock time is reduced considerably. Nevertheless, it still may be that processor 0 has more work to perform so that processor 1 spends much time waiting for processor 0 to finish up a particular task. It is easy to see that, in this case, the scaling will, in general, not be linear because processor 1 is not performing an equal share of the work. 

Wu Y.S., Zhang K., et al. An efficient parallel-computing scheme for modeling noni-sothermal multiphase flow and multicomponent transport in porous and fractured media. 2002 Advances in Water Resources 25 243-261. 

Zhang K., Doughty C., et al Efficient parallel simulation of C02 geologic sequestration in saline aquifers. In Proceeding of the 2007 SPE reservoir simulation synposium, Texas(106026). 2007 SPE. 

There have also been many theoretical developments that have extended the applicability and functionality of the DMRG method for quantum chemistry. Some have been algorithmic nature, for example, efficient parallel algorithms... 

From these studies it has been shown that for a successful and efficient parallel kinetic resolution the following guidelines need to be adhered to a) derivatisation with two complementary chiral reagents have to occur without mutual interference [17] b) both reactions need to occur with similar but preferably equal rate and have complementary stereocontrol and c) afford distinct and easily separable products. 

Although a theoretical approach has been desecrated as to how one can apply the generalized coupled master equations to deal with ultrafast radiationless transitions taking place in molecular systems, there are several problems and limitations to the approach. For example, the number of the vibrational modes is limited to less than six for numerical calculations. This is simply just because of the limitation of the computational resources. If the efficient parallelization can be realized to the generalized coupled master equations, the limitation of the number of the modes can be relaxed. In the present approach, the Markov approximation to the interaction between the molecule and the heat bath mode has been employed. If the time scale of the ultrashort measurements becomes close to the characteristic time of the correlation time of the heat bath mode, the Markov approximation cannot be applicable. In this case, the so-called non-Markov treatment should be used. This, in turn, leads to a more computationally demanding task. Thus, it is desirable to develop a new theoretical approach that allows a more efficient algorithm for the computation of the non-Markov kernels. Another problem is related to the modeling of the interaction between the molecule and the heat bath mode. In our model, the heat bath mode is treated as... 

Personal computing as defined here will begin to impact the author s field of application as the PC becomes integrated into the UNIX network environment. Aspects of this have already been achieved, but full realization of chemists goals in this area awaits availability of efficient parallelized supercomputers with speeds 10 to 1000 times that of today s largest computing systems. This will enable us to interactively model protein molecules and other complex systems in realistic chemical environments. 

Let me first describe the situation without umklapp (g3 = 0). For positive j, the fixed points are at g = 0 and 2 = 82 i/2. For negative x, the g s scale to the strong-coupling values f = -2, = 82 i/2 — 1. This can be understood in the following way. gi is known from boson-ization to be associated with the spin excitations. If gx is positive, the electrons will make use of the Pauli principle to avoid one another efficiently. Parallel-spin (triplet) correlations will be favored on which has no effect. For gx < 0, the opposite situation prevails. The electrons want to take advantage of the interaction and will thus prefer singlet correlations in which there is nothing to prevent an intimate contact. A gap opens up in the spin excitations. 

We now reexamine message passing as it pertains to software development and the interface of application software and library software. Compiler-managed parallelism is not yet ready for prime time. This means that efficient parallel programs are usually coded by hand, typically using point-to-point message passing and simple collective communications. There are many problems associated with this approach. 

The development and efficient implementation of a parallel direct SCF Hartree-Fock algorithm, with gradients and random phase approximation solutions, are described by Feyereisen and Kendall, who discussed details of the structure of the parallel version of DISCO. Preliminary results for calculations using the Intel-Delta parallel computer system were reported. The data showed that the algorithms were efficiently parallelized and that throughput of a one-processor Cray X-MP was reached with about 16 nodes on the Intel-Delta. The data also indicated that sequential code, which was not a bottleneck on traditional supercomputers, became time-critical on parallel computers. 

To date the most efficient parallel SCF algorithms have been based on replication within each processor of several 0 N ) matrices, limiting the maximum calculation size and forcing an unacceptably low ratio of processors to memory. This restriction led to increased activity in the development of distributed-data schemes, some of which we now consider. 

P. Hammarlund, Techniques for Efficient Parallel Scientific Computing, PhD thesis. Royal Institute of Technology, Stockholm, Sweden, 1996. 

R. Bhupathiraju, P. T. Cummings, and H. D. Cochran, Mol. Phys., 88, 1665 (1996). An Efficient Parallel Algorithm for Non-Equilibrium Simulations of Very Large Systems in Planar Couette Flow. 

ToF systems are well suited to the pulsed nature of SIMS and MALDI measurements, and combine many characteristics that are advantageous to imaging experiments. High detection efficiency, parallel detection, and a wide m/z detection range have made them ubiquitous for MALDI and SIMS-MSI. Recent improvements in measurement speed have particularly benefited MSI experiments. Modem commercially available MALDI-ToF instruments are equipped with solid-state lasers that allow acquisitions at up to 2 kHz. Such speed is a prerequisite for the development of higher spatial resolution analysis and three-dimensional imaging or for the analysis of large animal/ patient series of tissues. 

Efficient parallel synthesis of larger numbers of single defined compounds plays an ever more important role in dmg research. While it is possible to manually synthesize mixtures of large numbers of compounds on a solid phase employing the spHt-and-mix method, in parallel single compound synthesis a number of reaction vessels equal to the number of desired final products has to be handled (at least in the final step of the synthetic sequence) [l]. This latter approach clearly benefits from, or even requires assistance of some sort of robotic equipment/ devices. 

A hybridizing probe in an SBH experiment has completely random access to the data set (sequence), locating matching sequence strings anywhere within the target. This random access allows efficient parallel data processing, with the potential for thousands of probe-DNA hybridizations to be conducted simultaneously on a probe-or DNA-array or many other ways. One approach is to hybridize one labeled probe or probe sets at a time with arrays of samples [see,e.g., 12,19). Alternately, a single labeled sample can be hybridized with an array of probes [see,... 

Landsberg, A.M., T. R. Young, and J.P. Boris. 1994. An efficient parallel method for solving flows in complex three-dimensional geometries. AIAA Paper No. 94-0413. 

---