scholarly journals A Divide and Conquer Strategy for Scaling Weather Simulations with Multiple Regions of Interest

2013 ◽  
Vol 21 (3-4) ◽  
pp. 93-107
Author(s):  
Preeti Malakar ◽  
Thomas George ◽  
Sameer Kumar ◽  
Rashmi Mittal ◽  
Vijay Natarajan ◽  
...  
Keyword(s):  
Performance Improvement ◽  
Performance Prediction ◽  
Parallel Execution ◽  
Regions Of Interest ◽  
Divide And Conquer ◽  
Processor Allocation ◽  
Simulation Domain ◽  
Multiple Regions ◽  
Blue Gene ◽  
Sequential Strategy

Accurate and timely prediction of weather phenomena, such as hurricanes and flash floods, require high-fidelity compute intensive simulations of multiple finer regions of interest within a coarse simulation domain. Current weather applications execute these nested simulations sequentially using all the available processors, which is sub-optimal due to their sub-linear scalability. In this work, we present a strategy for parallel execution of multiple nested domain simulations based on partitioning the 2-D processor grid into disjoint rectangular regions associated with each domain. We propose a novel combination of performance prediction, processor allocation methods and topology-aware mapping of the regions on torus interconnects. Experiments on IBM Blue Gene systems using WRF show that the proposed strategies result in performance improvement of up to 33% with topology-oblivious mapping and up to additional 7% with topology-aware mapping over the default sequential strategy.

Reactive Molecular Dynamics Simulations, Data Analytics and Visualization

2015 ◽  
Vol 1756 ◽  
Author(s):  
Priya Vashishta ◽  
Rajiv K. Kalia ◽  
Aiichiro Nakano ◽  
Ying Li ◽  
Ken-ichi Nomura ◽  
...  
Keyword(s):  
Molecular Dynamics ◽  
Hydrogen Production ◽  
Density Functional ◽  
Silica Surface ◽  
Hydrogen Gas ◽  
Divide And Conquer ◽  
List Type ◽  
Reactive Molecular Dynamics ◽  
Production Of Hydrogen ◽  
Blue Gene

ABSTRACTMultimillion-atom reactive molecular dynamics (RMD) and large quantum molecular dynamics (QMD) simulations are used to investigate structural and dynamical correlations under highly nonequilibrium conditions and reactive processes in nanostructured materials under extreme conditions. This paper discusses four simulations:1.RMD simulations of heated aluminum nanoparticles have been performed to study the fast oxidation reaction processes of the core (aluminum)-shell (alumina) nanoparticles and small complexes.2.Cavitation bubbles readily occur in fluids subjected to rapid changes in pressure. We have used billion-atom RMD simulations on a 163,840-processor Blue Gene/P supercomputer to investigate chemical and mechanical damages caused by shock-induced collapse of nanobubbles in water near silica surface. Collapse of an empty nanobubble generates high-speed nanojet, resulting in the formation of a pit on the surface. The gas-filled bubbles undergo partial collapse and consequently the damage on the silica surface is mitigated.3.Our QMD simulation reveals rapid hydrogen production from water by an Al superatom. We have found a low activation-barrier mechanism, in which a pair of Lewis acid and base sites on the Aln surface preferentially catalyzes hydrogen production.4.We have introduced an extension of the divide-and-conquer (DC) algorithmic paradigm called divide-conquer-recombine (DCR) to perform large QMD simulations on massively parallel supercomputers, in which interatomic forces are computed quantum mechanically in the framework of density functional theory (DFT). A benchmark test on an IBM Blue Gene/Q computer exhibits an isogranular parallel efficiency of 0.984 on 786,432 cores for a 50.3 million-atom SiC system. As a test of production runs, LDC-DFT-based QMD simulation involving 16,661 atoms was performed on the Blue Gene/Q to study on-demand production of hydrogen gas from water using LiAl alloy particles.

scholarly journals Parallel MapReduce: Maximizing Cloud Resource Utilization and Performance Improvement Using Parallel Execution Strategies

2018 ◽  
Vol 2018 ◽  
pp. 1-17
Author(s):  
Ahmed Abdulhakim Al-Absi ◽  
Najeeb Abbas Al-Sammarraie ◽  
Wael Mohamed Shaher Yafooz ◽  
Dae-Ki Kang
Keyword(s):  
Cloud Computing ◽  
Performance Improvement ◽  
Biomedical Applications ◽  
Large Data ◽  
Cost Effective ◽  
Parallel Execution ◽  
Computing Platform ◽  
Execution Strategy ◽  
Large Data Analysis ◽  
Cloud Resources

MapReduce is the preferred cloud computing framework used in large data analysis and application processing. MapReduce frameworks currently in place suffer performance degradation due to the adoption of sequential processing approaches with little modification and thus exhibit underutilization of cloud resources. To overcome this drawback and reduce costs, we introduce a Parallel MapReduce (PMR) framework in this paper. We design a novel parallel execution strategy of Map and Reduce worker nodes. Our strategy enables further performance improvement and efficient utilization of cloud resources execution of Map and Reduce functions to utilize multicore environments available with computing nodes. We explain in detail makespan modeling and working principle of the PMR framework in the paper. Performance of PMR is compared with Hadoop through experiments considering three biomedical applications. Experiments conducted for BLAST, CAP3, and DeepBind biomedical applications report makespan time reduction of 38.92%, 18.00%, and 34.62% considering the PMR framework against Hadoop framework. Experiments' results prove that the PMR cloud computing platform proposed is robust, cost-effective, and scalable, which sufficiently supports diverse applications on public and private cloud platforms. Consequently, overall presentation and results indicate that there is good matching between theoretical makespan modeling presented and experimental values investigated.

Towards an integrated study of urban CO2 emissions

2021 ◽  
Author(s):  
Lukas Pilz ◽  
Sanam Vardag ◽  
Ralph Kleinschek ◽  
Samuel Hammer ◽  
André Butz
Keyword(s):  
Performance Improvement ◽  
Urban Areas ◽  
Performance Comparison ◽  
Point Sources ◽  
Performance Ratio ◽  
Weather Research And Forecasting ◽  
Cavity Ringdown ◽  
Simulation Domain ◽  
Integrated Study

<p>High-resolution monitoring is the basis for CO<sub>2</sub> emissions tracking and attribution in urban areas. This work is an important step towards an integrated urban CO<sub>2</sub> emissions monitoring system. Three middle-cost nondispersive infrared (NDIR) sensors of 500€ to 3000€ are characterised. Furthermore, CO<sub>2</sub> emissions of large, regional point sources are simulated to analyse their effect on these sensors’ signals.</p><p>The three sensors are Vaisala GMP343, Senseair HPP3 and SmartGas FlowEvo CO<sub>2</sub>. Their analysis and characterisation is achieved by co-locating them with a Picarro G2401 cavity ringdown spectrometer for 40 days. Co-locating different middle-cost sensors is novel and enables a direct performance comparison. While the HPP3 is the only one to reach a 1 min mean standard deviation under 1 ppm, the GMP343 is the most linear and stable with a drift of 0.03(2) × 10<sup>−1</sup> ppm per day and the SmartGas sensor provides the best price-to-performance ratio. For all sensors, precisions (the 1 min mean error’s lower bound) of under 0.8 ppm are determined. In general, temperature stabilisation turns out to be one of the most promising avenues of performance improvement for all sensors.</p><p>The sensors’ in-situ measurements are combined with meso-scale meteorological simulations for the Rhine-Neckar region using the Weather Research and Forecasting model (WRF). In two case-studies, simulated excess CO<sub>2</sub> due to large, regional point sources and measured CO<sub>2</sub> concentration are compared. Both simulations show qualitative agreement with the measurements. The differences between measurements and simulation, however, highlight aspects to be refined. These include increasing the horizontal and vertical resolution of the simulation domain as well improving as the parametrisation of the planetary and urban boundary layer.</p>

Sign in / Sign up

Export Citation Format

Share Document