In order to create an efficient system for modeling giant reservoir models, there is a whole range of technical challenges which have to be addressed. In this paper we concentrate on the topic of parallel scalability of complex computing systems like multi-CPU clusters and workstations with GPU processing cards. For multi-CPU distributed memory computing systems, it is shown that about 10 times improvement in parallel performance can be achieved if a new so called "hybrid?? approach is used. In this "hybrid?? approach, the usual MPI synchronization between the cluster nodes is being interleaved with a shared memory system thread based synchronization at the node level. It is demonstrated, that for some "black oil?? models of real oil and gas fields, the parallel acceleration factor can exceed 1300 times for 4096 CPU cores. Even for the extreme example of a giant full field model containing over 14,000 production and injection wells, it is shown that a parallel acceleration of over 350 times can be achieved. For CPU-GPU, and CPU-CPU based systems, we compare the parallel performance of simple iterative and realistic pre-conditioner based algorithms typically used in oil and gas simulations. Hardware systems equipped with AMD FirePro, nVidia TESLA and 16-core dual Intel Xeon E2580 systems are compared in this study.
El-Amin, M.F. (King Abdullah University of Science and Technology) | Sun, Shuyu (King Abdullah University of Science and Technology) | Salama, Amgad (King Abdullah University of Science and Technology)
Geological storage of anthropogenic CO2 emissions in deep saline aquifers has recently received tremendous attention in the scientific literature. Injected CO2 plume buoyantly accumulates at the top part of the deep aquifer under a sealing cap rock, and some concern that the high-pressure CO2 could breach the seal rock. However, CO2 will diffuse into the brine underneath and generate a slightly denser fluid that may induce instability and convective mixing. Onset times of instability and convective mixing performance depend on the physical properties of the rock and fluids, such as permeability and density contrast. The novel idea is to adding nanoparticles to the injected CO2 to increase density contrast between the CO2-rich brine and the underlying resident brine and, consequently, decrease onset time of instability and increase convective mixing.
As far as it goes, only few works address the issues related to mathematical and numerical modeling aspects of the nanoparticles transport phenomena in CO2 storages. In the current work, we will present mathematical models to describe the nanoparticles transport carried by injected CO2 in porous media. Buoyancy and capillary forces as well as Brownian diffusion are important to be considered in the model. IMplicit Pressure Explicit Saturation-Concentration (IMPESC) scheme is used and a numerical simulator is developed to simulate the nanoparticles transport in CO2 storages.
ResInsight is an open source project developing a cross-platform 3D visualization application for reservoir models and simulations. The main motivation of the development work is to improve key aspects of the human interaction when working with large scale reservoir models. The paper illustrates visualizations of properties, faults and wells designed for efficient interpretation of reservoir simulation data. To be highly responsive during visualization of large datasets, the system exploits multi-core CPUs and GPUs by a new core visualization library and a limited-intrusive multi-threading approach. Powerful and flexible mechanisms for result manipulation and numerical computations are established based on GNU Octave. Derived results can be returned to ResInsight for further handling and visualization. Eventually, derived and computed properties can be directly exported to form input to new simulations, thereby reducing the time needed for simulation cycles and parameter studies.
ResInsight is an emerging application for 3D visualization of reservoir models and simulations. Major objectives of the development work were to build a dedicated application with tailored functionality and effective visualizations. To provide the desired efficiency, challenges concern both informative visualizations as well as the performance of the application.
Figure 1 shows the overall system architecture for ResInsight with the four underlying submodules that have served as building blocks during the development work. The Custom Visualization Core (CVC) is a modern OpenGL based class library developed by Ceetron for flexible and efficient development of custom 3D visualization applications. ResInsight also uses a part of the Ensemble based Reservoir Tool (ERT) developed by Statoil for reading ECLIPSE files . Furthermore, Qt is used as a powerful framework for developing the ResInsight user interface with a cross-platform capability . GNU Octave is a high-level interpreted language for numerical computations  which is used to extend the functionality of ResInsight. The distinct color in Figure 1 signals that ResInsight communicates with GNU Octave as a separate program.
ResInsight is cross-platform, open software  to encourage a broad usage and wide cooperation and thereby enhance the solutions, ensure robustness, and increase productivity. The project began just over a year ago and is expected to further beyond 2012.