scholarly journals The GPU version of LICOM3 under HIP framework and its large-scale application

2020 ◽  
Author(s):  
Pengfei Wang ◽  
Jinrong Jiang ◽  
Pengfei Lin ◽  
Mengrong Ding ◽  
Junlin Wei ◽  
...  
Keyword(s):  
Graphics Processing Units ◽  
General Circulation ◽  
High Performance ◽  
Large Scale ◽  
Data Transfer ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Model ◽  
Global Ocean ◽  
Almost All

Abstract. A high-resolution (1/20°) global ocean general circulation model with Graphics processing units (GPUs) code implementations is developed based on the LASG/IAP Climate system Ocean Model version 3 (LICOM3) under Heterogeneous-compute Interface for Portability (HIP) framework. The dynamic core and physics package of LICOM3 are both ported to the GPU, and 3-dimensional parallelization is applied. The HIP version of the LICOM3 (LICOM3-HIP) is 42 times faster than what the same number of CPU cores dose, when 384 AMD GPUs and CPU cores are used. The LICOM3-HIP has excellent scalability; it can still obtain speedup of more than four on 9216 GPUs comparing to 384 GPUs. In this phase, we successfully performed a test of 1/20° LICOM3-HIP using 6550 nodes and 26200 GPUs, and at the grand scale, the model’s time to solution can still obtain an increasing, about 2.72 simulated years per day (SYPD). The high performance was due to putting almost all of computation processes inside GPUs, and thus greatly reduces the time cost of data transfer between CPUs and GPUs. At the same time, a 14-year spin-up integration following the phase 2 of Ocean Model Intercomparison Project (OMIP-2) protocol of surface forcing has been conducted, and the preliminary results have been evaluated. We found that the model results have little differences from the CPU version. Further comparison with observations and lower-resolution LICOM3 results suggests that the 1/20° LICOM3-HIP can not only reproduce the observations, but also produce much smaller scale activities, such as submesoscale eddies and frontal scales structures.

scholarly journals The GPU version of LASG/IAP Climate System Ocean Model version 3 (LICOM3) under the heterogeneous-compute interface for portability (HIP) framework and its large-scale application

2021 ◽  
Vol 14 (5) ◽  
pp. 2781-2799
Author(s):  
Pengfei Wang ◽  
Jinrong Jiang ◽  
Pengfei Lin ◽  
Mengrong Ding ◽  
Junlin Wei ◽  
...  
Keyword(s):  
General Circulation ◽  
Large Scale ◽  
Vertical Direction ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Model ◽  
Climate System ◽  
Global Ocean ◽  
Processing Unit ◽  
Model Version

Abstract. A high-resolution (1/20∘) global ocean general circulation model with graphics processing unit (GPU) code implementations is developed based on the LASG/IAP Climate System Ocean Model version 3 (LICOM3) under a heterogeneous-compute interface for portability (HIP) framework. The dynamic core and physics package of LICOM3 are both ported to the GPU, and three-dimensional parallelization (also partitioned in the vertical direction) is applied. The HIP version of LICOM3 (LICOM3-HIP) is 42 times faster than the same number of CPU cores when 384 AMD GPUs and CPU cores are used. LICOM3-HIP has excellent scalability; it can still obtain a speedup of more than 4 on 9216 GPUs compared to 384 GPUs. In this phase, we successfully performed a test of 1/20∘ LICOM3-HIP using 6550 nodes and 26 200 GPUs, and on a large scale, the model's speed was increased to approximately 2.72 simulated years per day (SYPD). By putting almost all the computation processes inside GPUs, the time cost of data transfer between CPUs and GPUs was reduced, resulting in high performance. Simultaneously, a 14-year spin-up integration following phase 2 of the Ocean Model Intercomparison Project (OMIP-2) protocol of surface forcing was performed, and preliminary results were evaluated. We found that the model results had little difference from the CPU version. Further comparison with observations and lower-resolution LICOM3 results suggests that the 1/20∘ LICOM3-HIP can reproduce the observations and produce many smaller-scale activities, such as submesoscale eddies and frontal-scale structures.

scholarly journals Can Lagrangian Tracking Simulate Tracer Spreading in a High-Resolution Ocean General Circulation Model?

2019 ◽  
Vol 49 (5) ◽  
pp. 1141-1157 ◽  
Author(s):  
Patrick Wagner ◽  
Siren Rühs ◽  
Franziska U. Schwarzkopf ◽  
Inga Monika Koszalka ◽  
Arne Biastoch
Keyword(s):  
General Circulation ◽  
Large Scale ◽  
Circulation Model ◽  
Eddy Diffusivity ◽  
Ocean General Circulation Model ◽  
Horizontal Resolution ◽  
Ocean Model ◽  
Time Step ◽  
Lagrangian Particles ◽  
Cape Basin

AbstractTo model tracer spreading in the ocean, Lagrangian simulations in an offline framework are a practical and efficient alternative to solving the advective–diffusive tracer equations online. Differences in both approaches raise the question of whether both methods are comparable. Lagrangian simulations usually use model output averaged in time, and trajectories are not subject to parameterized subgrid diffusion, which is included in the advection–diffusion equations of ocean models. Previous studies focused on diffusivity estimates in idealized models but could show that both methods yield similar results as long as the deformations-scale dynamics are resolved and a sufficient amount of Lagrangian particles is used. This study compares spreading of an Eulerian tracer simulated online and a cloud of Lagrangian particles simulated offline with velocities from the same ocean model. We use a global, eddy-resolving ocean model featuring 1/20° horizontal resolution in the Agulhas region around South Africa. Tracer and particles were released at one time step in the Cape Basin and below the mixed layer and integrated for 3 years. Large-scale diagnostics, like mean pathways of floats and tracer, are almost identical and 1D horizontal distributions show no significant differences. Differences in vertical distributions, seen in a reduced vertical spreading and downward displacement of particles, are due to the combined effect of unresolved subdaily variability of the vertical velocities and the spatial variation of vertical diffusivity. This, in turn, has a small impact on the horizontal spreading behavior. The estimates of eddy diffusivity from particles and tracer yield comparable results of about 4000 m2 s−1 in the Cape Basin.

scholarly journals High frequency fluctuations in the heat content of an ocean general circulation model

2012 ◽  
Vol 9 (1) ◽  
pp. 25-61
Author(s):  
A. M. Huerta-Casas ◽  
D. J. Webb
Keyword(s):  
General Circulation ◽  
Heat Budget ◽  
Experimental Studies ◽  
Heat Content ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Model ◽  
Ocean Dynamics ◽  
Global Ocean ◽  
Model Studies

Abstract. The transport and storage of heat by the ocean is of crucial importance because of its effect on ocean dynamics and its impact on the atmosphere, climate and climate change. Unfortunately, limits to the amount of data that can be collected and stored means that many experimental and modelling studies of the heat budget have to make use of mean datasets where the effects of short term fluctuations are lost. In this paper we investigate the magnitude of the resulting errors making use of data from OCCAM, a high resolution global ocean model. The model carries out a proper heat balance every timestep so any imbalances that are found in the analysis must result from the use of mean fields. The study concentrates on two areas of the ocean affecting the El Nino. The first is the region of tropical instability waves north of the equator. The second is in the upwelling region along the equator. It is shown that in both cases, processes with a period of less than five days can have a significant impact on the heat budget. Thus analyses using data averaged over five days or more are likely to have significant errors. It is also shown that if a series of instantaneous values is available, reasonable estimates can be made of the size of the errors. In model studies such values are available in the form of the datasets used to restart the model. In experimental studies they may be in the form of individual unaveraged observations.

scholarly journals Comparison of ocean heat content from two eddy-resolving hindcast simulations with OFES1 and OFES2

2021 ◽  
Author(s):  
Fanglou Liao ◽  
Xiao Hua Wang ◽  
Zhiqiang Liu
Keyword(s):  
Heat Transport ◽  
General Circulation ◽  
Large Scale ◽  
Potential Temperature ◽  
Heat Content ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Heat Content ◽  
Global Ocean ◽  
Spatial Distributions

Abstract. The ocean heat content (OHC) estimates from high-resolution hindcast simulations from the Ocean General Circulation Model for the Earth Simulator Version 1 (OFES1) and Version 2 (OFES2), and a global objective analysis of subsurface temperature observations (EN4.2.1) were compared. There was an OHC increase in most of the global ocean over a 57-year period, mainly a result of vertical displacements of neutral density surfaces. However, we found substantial differences in the temporal and meridional distributions of the OHC between the two OFES hindcasts. The spatial distributions of potential-temperature change also differed significantly, especially in the Atlantic Ocean. The spatial distributions of the time-averaged surface heat flux and heat transport from the OFES1 and OFES2 were highly correlated, but differences could be seen. However, these differences, more specifically in the heat transport, were only partially responsible for the OHC differences. The marked OHC differences may arise from the different vertical mixing schemes and may impact the large-scale pressure field, and thus the geostrophic current. The work here should be a useful reference for future OFES users.

scholarly journals Water isotope variations in the global ocean model MPI-OM

2012 ◽  
Vol 5 (3) ◽  
pp. 809-818 ◽  
Author(s):  
X. Xu ◽  
M. Werner ◽  
M. Butzin ◽  
G. Lohmann
Keyword(s):  
Surface Waters ◽  
General Circulation ◽  
Large Scale ◽  
Circulation Model ◽  
Deep Ocean ◽  
Ocean Surface ◽  
Ocean Model ◽  
Global Ocean ◽  
Stable Water Isotopes ◽  
Tropical Regions

Abstract. The stable water isotopes H218O and HDO are incorporated as passive tracers into the oceanic general circulation model MPI-OM, and a control simulation under present-day climate conditions is analyzed in detail. Both δ18O and δD distributions at the ocean surface and deep ocean are generally consistent with available observations on the large scale. The modelled δD-δ 18O relations in surface waters slightly deviates from the slope of the global meteoric water line in most basins, and a much steeper slope is detected in Arctic Oceans. The simulated deuterium excess of ocean surface waters shows small variations between 80° S and 55° N, and a strong decrease north of 55° N. The model is also able to capture the quasi-linear relationship between δ18O and salinity S, as well as δD and S, as seen in observational data. Both in the model results and observations, the surface δ−S relations show a steeper slope in extra-tropical regions than in tropical regions, which indicates relatively more addition of isotopically depleted water at high latitudes.

scholarly journals Formulation of a new explicit tidal scheme in ocean general circulation model

2022 ◽  
Author(s):  
Jiangbo Jin ◽  
Run Guo ◽  
Minghua Zhang ◽  
Guangqing Zhou ◽  
Qingcun Zeng
Keyword(s):  
Sea Level ◽  
General Circulation ◽  
Thermohaline Circulation ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Model ◽  
Tidal Mixing ◽  
Global Ocean ◽  
The North ◽  
Tidal Constituents

Abstract. Tides play an important role in ocean energy transfer and mixing, and provide major energy for maintaining thermohaline circulation. This study proposes a new explicit tidal scheme and assesses its performance in a global ocean model. Instead of using empirical specifications of tidal amplitudes and frequencies, the new scheme directly uses the positions of the Moon and Sun in a global ocean model to incorporate tides. Compared with the traditional method that has specified tidal constituents, the new scheme can better simulate the diurnal and spatial characteristics of the tidal potential of spring and neap tides as well as the spatial patterns and magnitudes of major tidal constituents (K1 and M2). It significantly reduces the total errors of eight tidal constituents (with the exception of N2 and Q1) in the traditional explicit tidal scheme. Relative to the control simulation without tides, both the new and traditional tidal schemes can lead to better dynamic sea level (DSL) simulation in the North Atlantic, reducing significant negative biases in this region. The new tidal scheme also shows smaller positive bias than the traditional scheme in the Southern Ocean. The new scheme is suited to calculate regional distributions of sea level height in addition to tidal mixing.

scholarly journals Water isotope variations in the global ocean model MPI-OM

2012 ◽  
Vol 5 (1) ◽  
pp. 277-307
Author(s):  
X. Xu ◽  
M. Werner ◽  
M. Butzin ◽  
G. Lohmann
Keyword(s):  
Surface Waters ◽  
General Circulation ◽  
Large Scale ◽  
Circulation Model ◽  
Deep Ocean ◽  
Ocean Surface ◽  
Ocean Model ◽  
Global Ocean ◽  
Stable Water Isotopes ◽  
Tropical Regions

Abstract. The stable water isotopes H218O and HDO are incorporated as passive tracers into the oceanic general circulation model MPI-OM, and a control simulation under present-day climate conditions is analyzed in detail. Both δ18O and δD distributions at the ocean surface and deep ocean are generally consistent with available observations on the large scale. The modelled δD-δ18O relations in surface waters slightly deviates from the slope of the global meteoric water line in most basins, and a much steeper slope is detected in Arctic Oceans. The simulated deuterium excess of ocean surface waters shows small variations between 80° S and 55° N, and a strong decrease north of 55° N. The model is also able to capture the quasi-linear relationship between δ18O and salinity S, as well as δD and S, as seen in observational data. Both in the model results and observations, the surface δ–S relations show a steeper slope in extra-tropical regions than in tropical regions, which indicates relatively more addition of isotopically depleted water at high latitudes.

scholarly journals High frequency fluctuations in the heat content of an ocean general circulation model

Ocean Science ◽  
2012 ◽  
Vol 8 (5) ◽  
pp. 813-825 ◽  
Author(s):  
A. M. Huerta-Casas ◽  
D. J. Webb
Keyword(s):  
General Circulation ◽  
Heat Budget ◽  
Experimental Studies ◽  
Heat Content ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Model ◽  
Ocean Dynamics ◽  
Global Ocean ◽  
Time Step

Abstract. The transport and storage of heat by the ocean is of crucial importance because of its effect on ocean dynamics and its impact on the atmosphere, climate and climate change. Unfortunately, limits to the amount of data that can be collected and stored mean that many experimental and modelling studies of the heat budget have to make use of mean datasets where the effects of short term fluctuations are lost. In this paper we investigate the magnitude of the resulting errors by making use of data from OCCAM, a high resolution global ocean model. The model carries out a proper heat balance every time step so any imbalances that are found in the analysis must result from the use of mean fields. The study concentrates on two areas of the ocean affecting the El Nino. The first is the region of tropical instability waves north of the Equator. The second is in the upwelling region along the Equator. It is shown that in both cases, processes with a period of less than five days can have a significant impact on the heat budget. Thus, analyses using data averaged over five days or more are likely to have significant errors. It is also shown that if a series of instantaneous values is available, reasonable estimates can be made of the size of the errors. In model studies, such values are available in the form of the datasets used to restart the model. In experimental studies they may be in the form of individual unaveraged observations.

scholarly journals Characteristics of the RAW-Filtered Leapfrog Time-Stepping Scheme in the Ocean General Circulation Model

2014 ◽  
Vol 142 (1) ◽  
pp. 434-447 ◽  
Author(s):  
Chih-Chieh Young ◽  
Yu-Chiao Liang ◽  
Yu-Heng Tseng ◽  
Chun-Hoe Chow
Keyword(s):  
General Circulation ◽  
Classical Model ◽  
Circulation Model ◽  
Ocean General Circulation Model ◽  
Ocean Model ◽  
Ocean Modeling ◽  
Global Ocean ◽  
Numerical Dissipation ◽  
Characteristic Analysis ◽  
Time Stepping

Abstract The Robert–Asselin–Williams (RAW) filtered leapfrog scheme is implemented and tested in the Taiwan multiscale community ocean model (TIMCOM). The characteristics of the RAW filter are carefully examined through two benchmark tests (the classical model problem-oscillation equation with further consideration of the dissipation effect, and the 1D linearized shallow-water equations). Particularly, the effect of the RAW filter upon the 2Δx wave instability due to spatial truncation errors is addressed. TIMCOM is then applied to simulate the coastally trapped internal Kelvin waves and global ocean circulations, showing the practical improvement over the Robert–Asselin (RA) filter in the short- and long-term model integrations. The large mean differences in some major current systems suggest the potential impacts on the oceanic instability where the numerical dissipation may interfere with the physical one. The characteristic analysis and model results here indicate the significant advantage of the RAW-filtered leapfrog time-stepping scheme for accurate ocean modeling.

Sign in / Sign up

Export Citation Format

Share Document