scholarly journals Accelerating the Global Nested Air Quality Prediction Modeling System (GNAQPMS) model on Intel Xeon Phi processors

2017 ◽  
Author(s):  
Hui Wang ◽  
Huansheng Chen ◽  
Qizhong Wu ◽  
Junming Lin ◽  
Xueshun Chen ◽  
...  
Keyword(s):  
Air Quality ◽  
Environmental Research ◽  
Quality Prediction ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Parallel Scalability ◽  
Air Quality Prediction ◽  
Air Quality Forecast ◽  
Parallel Mode ◽  
Intel Xeon

Abstract. The GNAQPMS model is the global version of the Nested Air Quality Prediction Modelling System (NAQPMS), which is a multi-scale chemical transport model used for air quality forecast and atmospheric environmental research. In this study, we present our work of porting and optimizing the GNAQPMS model on the second generation Intel Xeon Phi processor codename “Knights Landing” (KNL). Compared with the first generation Xeon Phi coprocessor, KNL introduced many new hardware features such as a bootable processor, high performance in-package memory and ISA compatibility with Intel Xeon processor. In particular, we described the five optimizations we applied to the key modules of GNAQPMS – CBM-Z gas chemistry, advection, convection and wet deposition. These optimizations work well on both the KNL 7250 processor as well as the Intel Xeon processor E5-2697 V4. They include: 1) updating the pure MPI parallel mode to hybrid parallel mode with MPI and OpenMP in emission, advection, convection and chemistry modules; 2) fully employ the 512-bit wide vector processing units (VPU) on the KNL platform; 3) reducing unnecessary memory access to improve caches efficiency; 4) reducing thread local storage (TLS) in CBM-Z gas phase chemistry module to improve its OpenMP performance; 5) changing global communication from interface-files writing/reading to using Message Passing Interface (MPI) functions to improve the performance and the parallel scalability. These optimizations improved GNAQPMS performance great. The same optimizations also work well for the Intel Xeon Broadwell processor, specifically, E5-2697v4. Compared with the baseline version of GNAQPMS, the optimized version is 3.34x faster on KNL and 2.39x faster on CPU. Furthermore, the optimized version on KNL runs at 26 % lower average power compare to CPU. Combining the performance and energy improvement, the KNL platform is 47% more efficient compare to the CPU platform. The optimizations also enables much further parallel scalability on both the CPU cluster and KNL cluster – scale to 40 CPU nodes and 30 KNL nodes, with a parallel efficiency of 70.4 % and 42.2 %, respectively.

scholarly journals GNAQPMS v1.1: accelerating the Global Nested Air Quality Prediction Modeling System (GNAQPMS) on Intel Xeon Phi processors

2017 ◽  
Vol 10 (8) ◽  
pp. 2891-2904 ◽  
Author(s):  
Hui Wang ◽  
Huansheng Chen ◽  
Qizhong Wu ◽  
Junmin Lin ◽  
Xueshun Chen ◽  
...  
Keyword(s):  
Air Quality ◽  
Gas Phase ◽  
Quality Prediction ◽  
Xeon Phi ◽  
Prediction Modeling ◽  
Gas Phase Chemistry ◽  
Air Quality Prediction ◽  
Parallel Mode ◽  
Phase Chemistry ◽  
Intel Xeon

Abstract. The Global Nested Air Quality Prediction Modeling System (GNAQPMS) is the global version of the Nested Air Quality Prediction Modeling System (NAQPMS), which is a multi-scale chemical transport model used for air quality forecast and atmospheric environmental research. In this study, we present the porting and optimisation of GNAQPMS on a second-generation Intel Xeon Phi processor, codenamed Knights Landing (KNL). Compared with the first-generation Xeon Phi coprocessor (codenamed Knights Corner, KNC), KNL has many new hardware features such as a bootable processor, high-performance in-package memory and ISA compatibility with Intel Xeon processors. In particular, we describe the five optimisations we applied to the key modules of GNAQPMS, including the CBM-Z gas-phase chemistry, advection, convection and wet deposition modules. These optimisations work well on both the KNL 7250 processor and the Intel Xeon E5-2697 V4 processor. They include (1) updating the pure Message Passing Interface (MPI) parallel mode to the hybrid parallel mode with MPI and OpenMP in the emission, advection, convection and gas-phase chemistry modules; (2) fully employing the 512 bit wide vector processing units (VPUs) on the KNL platform; (3) reducing unnecessary memory access to improve cache efficiency; (4) reducing the thread local storage (TLS) in the CBM-Z gas-phase chemistry module to improve its OpenMP performance; and (5) changing the global communication from writing/reading interface files to MPI functions to improve the performance and the parallel scalability. These optimisations greatly improved the GNAQPMS performance. The same optimisations also work well for the Intel Xeon Broadwell processor, specifically E5-2697 v4. Compared with the baseline version of GNAQPMS, the optimised version was 3.51 × faster on KNL and 2.77 × faster on the CPU. Moreover, the optimised version ran at 26 % lower average power on KNL than on the CPU. With the combined performance and energy improvement, the KNL platform was 37.5 % more efficient on power consumption compared with the CPU platform. The optimisations also enabled much further parallel scalability on both the CPU cluster and the KNL cluster scaled to 40 CPU nodes and 30 KNL nodes, with a parallel efficiency of 70.4 and 42.2 %, respectively.

Urbanization of Weather Forecast, Air-Quality Prediction and Climate Scenarios - Project URBI PRAGENSI

2021 ◽  
Author(s):  
Tomas Halenka ◽  
Michal Belda ◽  
Peter Huszar ◽  
Jan Karlicky ◽  
Tereza Novakova
Keyword(s):  
Air Quality ◽  
Weather Prediction ◽  
Regional Climate ◽  
Weather Forecast ◽  
Quality Prediction ◽  
Climate Scenarios ◽  
Air Quality Prediction ◽  
Air Quality Forecast ◽  
The City ◽  
Very High

<p>The ratio of population living in cities is growing and this is especially true for the largest ones, megacities. However, even smaller cities like the City of Prague  (about 1.5 M) can suffer significantly and the night time temperature difference under summer heat wave can achieve more than 5°C. To assess the impact of cities and urban structures on weather, climate and air-quality, modelling approach is commonly used and the inclusion of urban parameterization in land-surface interactions is of primary importance to capture the urban effects properly. This is especially important when going to higher resolution, which is common trend in operational weather forecast, air-quality prediction as well as regional climate modeling. This represents the rapidly developing research, motivated by specific risks in urban environment, with strong impacts on vulnerable communities there, leading to the tools to assess properly impacts within the cities and the effectiveness of adaptation and mitigation options applied there by the city authorities. Under the action towards the Smart Cities and within the framework for developing adequate climate services, such supporting tools for decission making are inevitable. It is valid not only for extreme heat waves impact prediction, but as well in air-quality forecast and in long term perspective in connection to climate change impacts assessment. This provides the background for the project within Operational Program Prague - The Pole of Growth “Urbanization of weather forecast, air-quality prediction and climate scenarios for Prague”, shortly URBI PRAGENSI.</p><p> </p><p>There are four main tasks within the project. First, urbanization of weather forecast, i.e. involving and testing the urban parameterization scheme in the weather prediction model can provide in very high resolution localized weather prediction and especially under the heat wave condition it can well capture the temperature differences in the city center with respect to the remote areas. There are applications, which can use such localized prediction for planning and decision making on e.g. public services for some specific groups of population in risks. Further, air-quality forecast based on such urbanized weather condition forecast can benefit from better estimates of temperature for chemical reactions, mixing height for dispersion conditions etc. Third, urbanized scenarios of climate change can provide better description of future conditions in the city for adaptation and mitigation options, moreover, in connection to urban heat island urbanized regional climate model in very high resolution is good tool for estimates of efficiency  of potential adaptation or mitigation measures which might be applied by the city administration. Last, but not least, microscale simulations using LES methods are supposed to be used for selected local hot-spots to solve them.</p>

scholarly journals Air Quality Prediction Using Supervised Machine Learning Technique

2021 ◽  
Vol 23 (08) ◽  
pp. 62-69
Author(s):  
D. Joshna ◽  
◽  
K. Madhurya ◽  
K. Srividya ◽  
K. Ramamohanarao ◽  
...  
Keyword(s):  
Air Quality ◽  
Supervised Machine Learning ◽  
Quality Prediction ◽  
Machine Learning Technique ◽  
Great Opportunity ◽  
Air Contamination ◽  
Human Wellbeing ◽  
Air Quality Prediction ◽  
Air Quality Forecast ◽  
Learning Technique

Generally, air contamination alludes to the arrival of different contamination into the air which are compromising the human wellbeing and planet also. The air contamination is the major hazardous horrendous to humankind at any point confronted. It causes major harm to creatures, plants and so forth, if this continues proceeding, the individuals will confront major circumstances in the forthcoming years. The significant toxins are from the vehicle and enterprises. In this way, to forestall this issue significant areas need to foresee the air quality from transport and ventures .In existing undertaking there are numerous hindrances. The venture is tied in with assessing the PM2.5 fixation by planning a photo based strategy. In any case photographic technique isn’t the only one adequate to compute PM2.5 since it contains just one of the grouping of toxins furthermore, it ascertains just PM2.5 so there are some passing up a great opportunity of the significant toxins and the data required for controlling the contamination .So along these lines we proposed the AI procedures by UI of GUI application. In this numerous dataset can be joined from the diverse source to shape a summed up dataset and different AI calculations are used to get the outcomes with the most extreme precision. From looking at different AI calculations we can get the best precision result. Our assessment gives the thorough manual to affectability assessment of model boundaries concerning generally speaking execution in forecast of air great contaminations through exactness computation. Furthermore to examine and think about the presentation of AI calculations from the dataset with assessment of GUI based UI air quality forecast by credits.

scholarly journals MILC Code Performance on High End CPU and GPU Supercomputer Clusters

2018 ◽  
Vol 175 ◽  
pp. 02009
Author(s):  
Carleton DeTar ◽  
Steven Gottlieb ◽  
Ruizi Li ◽  
Doug Toussaint
Keyword(s):  
Conjugate Gradient ◽  
Memory Hierarchy ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Code Performance ◽  
Recent Developments ◽  
Knights Landing ◽  
Many Core ◽  
Intel Xeon

With recent developments in parallel supercomputing architecture, many core, multi-core, and GPU processors are now commonplace, resulting in more levels of parallelism, memory hierarchy, and programming complexity. It has been necessary to adapt the MILC code to these new processors starting with NVIDIA GPUs, and more recently, the Intel Xeon Phi processors. We report on our efforts to port and optimize our code for the Intel Knights Landing architecture. We consider performance of the MILC code with MPI and OpenMP, and optimizations with QOPQDP and QPhiX. For the latter approach, we concentrate on the staggered conjugate gradient and gauge force. We also consider performance on recent NVIDIA GPUs using the QUDA library.

Performance Evaluation of NAS Parallel Benchmarks on Intel Xeon Phi

2013 ◽  
Author(s):  
Arunmoezhi Ramachandran ◽  
Jerome Vienne ◽  
Rob Van Der Wijngaart ◽  
Lars Koesterke ◽  
Ilya Sharapov
Keyword(s):  
Performance Evaluation ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Intel Xeon
Sign in / Sign up

Export Citation Format

Share Document