High-performance Simulation Research Articles

Study on High-Performance Simulation Computer for Large-Scale System of Systems Simulation

The high-performance parallel computing (HPPC) has a better overall performance and higher productivity, for a generical large-scale army equipment system of systems (AESoS) simulation, and the runtime efficiency can be multiplied several tenfold to several hundredfold. The requirement analysis of simulation framework of AESoS based on HPPC was proposed. After the simulation framework of AESoS based on HPPC and its key techniques were discussed, the simulation framework of AESoS Based HPPC was designed. it is of great significance to offer certain references for the engineering application in the simulation fields of AESoS based on HPPC.

FDTD simulation of the acoustic interaction between a cavity and an adaptive Helmholtz resonator

Helmholtz resonators are used to manipulate the eigen behaviour of the acoustic fluid in a cavity. Using the analogy to a mass damper, it is possible to tune the resonators' eigen frequency by handling the resonator volume as spring and the resonator neck as mass of a vibration absorber. Simulating the transient interaction of the HR with the acoustic sound field in a cavity in terms of active sound control, a high simulation performance is demanded. This can be provided by the finite difference time domain method. The simulations are performed with respect to the LOEWE-Zentrum AdRIA acoustics demonstrator. It depicts the abstraction of a real room. Therefore, it consists of six sound reflecting walls that are implemented through acoustic impedances in the time domain. The semi-passive system is implemented using MATLAB/Simulink. In addition, the simulations are validated applying an adaptive Helmholtz resonator inside the acoustic fluid of the acoustics demonstrator.

A perspective on coupled multiscale simulation and validation in nuclear materials

Abstract

High Performance Simulation of Environmental Tracers in Heterogeneous Domains

In this study, we use PFLOTRAN, a highly scalable, parallel, flow, and reactive transport code to simulate the concentrations of 3H, 3He, CFC-11, CFC-12, CFC-113, SF6, 39Ar, and the mean groundwater age in heterogeneous fields on grids with an excess of 10 million nodes. We utilize this computational platform to simulate the concentration of multiple tracers in high-resolution, heterogeneous 2D and 3D domains, and calculate tracer-derived ages. Tracer-derived ages show systematic biases toward younger ages when the groundwater age distribution contains water older than the maximum tracer age. The deviation of the tracer-derived age distribution from the true groundwater age distribution increases with increasing heterogeneity of the system. However, the effect of heterogeneity is diminished as the mean travel time gets closer to the tracer age limit. Age distributions in 3D domains differ significantly from 2D domains. 3D simulations show decreased mean age, and less variance in age distribution for identical heterogeneity statistics. High-performance computing allows for investigation of tracer and groundwater age systematics in high-resolution domains, providing a platform for understanding and utilizing environmental tracer and groundwater age information in heterogeneous 3D systems.

Review of RT-LAB and Steps Involved for Implementation of a Simulink Model from MATLAB to REAL-TIME

Abstract In recent days, every researcher wants to develop his/her model in real-time. Simulation tools have been widely used for the design and improvement of electrical systems since the mid-twentieth century. The evolution of simulation tools has progressed in step with the evolution of computing technologies. Now a days, computing technologies have improved dramatically in performance and become widely available at a steadily decreasing cost. Consequently, simulation tools have also seen dramatic performance gains and steady cost decreases. Researchers and engineers now have access to affordable, high-performance simulation tools that were previously too cost prohibitive, except for the largest manufacturers and utilities.This article has introduced a specific class of digital simulator known as a real-time simulator by answering the questions “What is real-time simulation?” “Why is it needed” and “How it works”. The latest trend in real-time simulation consists of exporting simulation models to FPGA. In this article, the steps involved for implementation of a model from MATLAB to REAL-TIME are provided in detail. The detailed real-time results are presented to support the feasibility of real-time digital simulator.

High-performance simulation and simulation methodologies

The realization of high-performance simulation necessitates sophisticated simulation experimentation and optimization; this often requires non-trivial amounts of computing power. Distributed computing techniques and systems found in areas such as high-performance computing (HPC), high-throughput computing (HTC), e-Infrastructures, grid and cloud computing can provide the required computing capacity for the execution of large and complex simulations. This extends the long tradition of adopting advances in distributed computing in simulation as evidenced by contributions from the parallel and distributed simulation community. There has arguably been a recent acceleration of innovation in distributed computing tools and techniques. This special issue presents the opportunity to showcase recent research that is assimilating these new advances in simulation. This special issue brings together a contemporary collection of work showcasing original research in the advancement of simulation theory and practice with distributed computing. This special issue has two parts. The first part (published in the preceding issue of the journal) included seven studies in high-performance simulation that support applications including the study of epidemics, social networks, urban mobility and real-time embedded and cyber-physical systems. This second part focuses on original research in high performance simulation that supports a range of methods including Discrete Event System Specification (DEVS), Petri nets and discrete event simulation (DES). Note that the manuscript by Bergero et al., which was submitted, reviewed and accepted for the special issue, was published in an earlier issue of SIMULATION as the author requested early publication.

Application and support for high-performance simulation

High-performance simulation that supports sophisticated simulation experimentation and optimization can require non-trivial amounts of computing power. Advanced distributed computing techniques and systems found in areas such as High-performance Computing (HPC), Highthroughput Computing (HTC), grid computing, cloud computing and e-Infrastructures are needed to provide effectively the computing power needed for the highperformance simulation of large and complex models. In simulation there has been a long tradition of translating and adopting advances in distributed computing, as shown by contributions from the parallel and distributed simulation community. This special issue brings together a contemporary collection of work showcasing original research in the advancement of simulation theory and practice with distributed computing. This special issue is divided into two parts. This first part focuses on research pertaining to high-performance simulation that supports a range of applications, including the study of epidemics, social networks, urban mobility and real-time embedded and cyber-physical systems. Compared to other simulation techniques, agent-based modeling and simulation (MS however, it is increasingly being used to study large-scale problems. Agent-based simulations present challenges for high-performance simulation, as they can be complex and computationally demanding, and it is therefore not surprising that this special issue includes several articles on the high-performance simulation of such systems.

Modeling of nitrate concentration in groundwater using artificial intelligence approach—a case study of Gaza coastal aquifer

Nitrate concentration in groundwater is influenced by complex and interrelated variables, leading to great difficulty during the modeling process. The objectives of this study are (1) to evaluate the performance of two artificial intelligence (AI) techniques, namely artificial neural networks and support vector machine, in modeling groundwater nitrate concentration using scant input data, as well as (2) to assess the effect of data clustering as a pre-modeling technique on the developed models' performance. The AI models were developed using data from 22 municipal wells of the Gaza coastal aquifer in Palestine from 2000 to 2010. Results indicated high simulation performance, with the correlation coefficient and the mean average percentage error of the best model reaching 0.996 and 7%, respectively. The variables that strongly influenced groundwater nitrate concentration were previous nitrate concentration, groundwater recharge, and on-ground nitrogen load of each land use land cover category in the well's vicinity. The results also demonstrated the merit of performing clustering of input data prior to the application of AI models. With their high performance and simplicity, the developed AI models can be effectively utilized to assess the effects of future management scenarios on groundwater nitrate concentration, leading to more reasonable groundwater resources management and decision-making.

Development of a coupled thermo-hydro-mechanical model in discontinuous media for carbon sequestration

Geomechanical alteration of porous media is generally ignored for most shallow subsurface applications, whereas carbon dioxide (CO2) injection, migration, and trapping in deep saline aquifers will be controlled by coupled multifluid flow, energy transfer, and geomechanical processes. The accurate assessment of the risks associated with potential leakage of injected CO2 and the design of effective injection systems require that we represent these coupled processes within numerical simulators. The objectives of this study were to develop a coupled thermo-hydro-mechanical model into a single software and to examine the coupling of thermal, hydrological, and geomechanical processes for simulation of CO2 injection into the subsurface for carbon sequestration. A numerical model was developed to couple nonisothermal multiphase hydrological and geomechanical processes for prediction of multiple interconnected processes for carbon sequestration in deep saline aquifers. The geomechanics model was based on the Rigid Body-Spring Model (RBSM), a discrete method for modeling discontinuous rock systems. Poisson’s effect that was often ignored by RBSM was considered in the model. The simulation of large-scale and long-term coupled processes in carbon capture and storage projects requires large memory and computational performance. The Global Array Toolkit was used to build the model to permit high-performance simulations of coupled processes. The model was used to simulate a case study with several scenarios to demonstrate the impacts of considering coupled processes and Poisson’s effect for the prediction of CO2 sequestration. As a demonstration of the coupled model, a conceptual 3D model was used to explain the double-lobe uplift pattern observed in the Krechba gas field at In Salah (Algeria), a site that demonstrated the success of a CO2 sequestration effort into a deep saline formation.

PARALLEL SIMULATION OF LARGE-SCALE ARTIFICIAL SOCIETY WITH GPU AS COPROCESSOR

Large-scale artificial societies with millions or billions of agents call for high-performance parallel simulation. Prevailing supercomputers with thousands of CPUs and GPUs make it possible to carry out such simulation. The key is to distribute large-scale agents to massive cores of CPUs and GPUs properly for parallel computing with efficient communication and synchronization. For simplicity and efficiency, a modified discrete event system specification (DEVS) is proposed for large-scale artificial society modeling and parallelism is exploited in agent models because similar agents usually share similar behaviors. Through phased synchronization, a two-tier parallel simulation engine is designed with support of MPI and OpenCL where GPU is used as coprocessor. One-sided communication is used for reflection of remote simulation objects and message passing between processes. A general kernel function prototype is elaborately designed and conditionally compiled for execution on both CPU and GPU. An artificial society for epidemic study is used to test the performance on a supercomputer with 1024 CPU cores and 1792 GPU cores. The speedup reaches 3512 for even 2 billion agents with GPU acceleration which is far over 701 when only CPUs are used. It turns out feasible for parallel simulation of large-scale artificial society with GPU as coprocessor.

Enabling Digital Earth simulation models using cloud computing or grid computing – two approaches supporting high-performance GIS simulation frameworks

Geospatial simulation models can help us understand the dynamic aspects of Digital Earth. To implement high-performance simulation models for complex geospatial problems, grid computing and cloud computing are two promising computational frameworks. This research compares the benefits and drawbacks of both in Web-based frameworks by testing a parallel Geographic Information System (GIS) simulation model (Schelling's residential segregation model). The parallel GIS simulation model was tested on XSEDE (a representative grid computing platform) and Amazon EC2 (a representative cloud computing platform). The test results demonstrate that cloud computing platforms can provide almost the same parallel computing capability as high-end grid computing frameworks. However, cloud computing resources are more accessible to individual scientists, easier to request and set up, and have more scalable software architecture for on-demand and dedicated Web services. These advantages may attract more geospatial scientists to utilize cloud computing for the development of Digital Earth simulation models in the future.

High performance computing and simulation: architectures, systems, algorithms, technologies, services, and applications

High performance computing and simulation: architectures, systems, algorithms, technologies, services, and applications

High-performance Computation-based Geographical Data Processing and Process Simulation

High-performance Computation-based Geographical Data Processing and Process Simulation

Cross-platform validation for a model of a low-cost Stewart platform

Low-cost motion systems have been proposed for certain training tasks that would otherwise be performed on high-performance full flight simulators. These systems have shorter stroke actuators, lower bandwidth, and higher motion noise. The influence of these characteristics on pilot perception and control behavior is unknown, and needs to be investigated. A possible approach to this would be to simulate a platform with limited capabilities with a high-end platform, and then remove the platform limitations one by one. The effects of these platform limitations on pilot behavior can then be investigated in isolation. In this paper, a model of a low-cost simulator was validated for simulation on a high-performance simulator. A dynamic model of the MPI Stewart platform was analyzed and compared with measurements of the baseline simulator response. Measurements for validation of the implementation of the model on the SIMONA Research Simulator showed that the dynamics of the MPI Stewart platform could be represented well in terms of dynamic range, time delay, and noise characteristics. The implementation of the model of the MPI Stewart platform will be used in experiments on the effects of these characteristics on pilot control behavior.

Using shared memory as a cache in high performance cellular automata water flow simulations

Graphics processors (GPU -- Graphic Processor Units) recently have gained a lot of interest as an efficient platform for general-purpose computation. Cellular Automata approach which is inherently parallel gives the opportunity to implement high performance simulations. This paper presents how shared memory in GPU can be used to improve performance for Cellular Automata models. In our previous works, we proposed algorithms for Cellular Automata model that use only a GPU global memory. Using a profiling tool, we found bottlenecks in our approach. We introduce modifications that takes an advantage of fast shared memory. The modified algorithm is presented in details, and the results of profiling and performance test are demonstrated. Our unique achievement is comparing the efficiency of the same algorithm working with a global and shared memory.

Three Tools for the Real-Time Simulation of Embodied Spiking Neural Networks Using GPUs

This paper presents a toolbox of solutions that enable the user to construct biologically-inspired spiking neural networks with tens of thousands of neurons and millions of connections that can be simulated in real time, visualized in 3D and connected to robots and other devices. NeMo is a high performance simulator that works with a variety of neural and oscillator models and performs parallel simulations on either GPUs or multi-core processors. SpikeStream is a visualization and analysis environment that works with NeMo and can construct networks, store them in a database and visualize their activity in 3D. The iSpike library provides biologically-inspired conversion between real data and spike representations to support work with robots, such as the iCub. Each of the tools described in this paper can be used independently with other software, and they also work well together.

Improved real‐time scheduling of periodic tasks on multiprocessors

SummaryThere is an increasing number of high‐performance periodic real‐time applications in areas such as control systems, autonomous robots and financial systems. This article presents a novel algorithm, called Notional Approximation for Balancing Load Residues (NABLR), for scheduling these applications on high‐performance computing resources. The algorithm utilizes a combination of task residual loads and runtime laxities to carefully plan task execution between two consecutive job arrivals, so that available resources can be fully utilized and avoid deadline misses as possible. The empirical study in our article presented at the 2011 International Conference on High Performance Computing and Simulation (HPCS) was further extended by including additional static task sets and a new adaptive task set generated by our motivating application in brain–machine interfaces, which simulates the control of movement of a prosthetic limb according to activities of input signals. Out of 25,000 task sets, NABLR can schedule up to 76% of the sets versus 43% by the best known efficient algorithm (named anticipating slack earliest deadline first until zero laxity [ASEDZL]), while incurring significantly smaller overheads than those of a known optimal algorithm (on average, 80% fewer preemptions, migrations, and 75% fewer scheduler invocations), and being comparable to those of suboptimal schedulers (within only 12% more preemptions/migrations). Additionally, the evaluation results show that NABLR completes more task instances when compared with ASEDZL, which yields a greater system output accuracy. Copyright © 2012 John Wiley & Sons, Ltd.

A reply

We thank Mercer for his letter in response to our article [1]. We have experienced similar difficulties and accept that problems arise from the linear course of events simulated by deductive models. Rather than being a problem of deductive modeling per se, however, we suggest that this requires physiological fidelity on the part of system designers, and deeper ‘architectural’ insight on the part of the controller(s) in order to use the system appropriately and effectively. In our opinion, the conduct of a complex scenario involving a five-member faculty is unaffordable. It is our belief that low-cost, high-fidelity deductive models will derive from high-performance simulators in future, and interface with low-cost manikins. Recently, we have demonstrated this approach successfully using tablet computers for virtual pacemaker teaching [2]. In both ‘deductive’ and ‘on the fly’ (instructor-driven) models the candidate is interacting with a physiological model. In the former, the model is a software program written by an engineering team, and in the latter a mental construct in the head of the manikin controller. Both models have problems with validity. The key advantages of a software model are that it can be based on published evidence, it can be subjected to peer review, and its limitations can be openly described. In contrast, the individual controlling the manikin introduces an unquantifiable risk of negative learning. We accept that further problem inherent to ‘deductive’ simulation is that scenario controllers do not necessarily want realism. For example, in the event of oxygen supply failure, arterial oxygen saturation often declines fairly slowly towards 90% but thereafter decreases rapidly. However, scenario controllers may simulate a gradual and linear saturation decline in order to give candidates time to implement their rescue. Whether or not this practice is a good idea, and indeed improves the learning experience, is an important question. Finally, we contend Mercer’s closing statement. In addition to faculty experience, simulated alterations should be conducted according to a valid evidence base. This is best achieved in a peer-reviewed physiological model.

Semantics and efficient simulation of accuracy-adaptive TLMs

In contrast to traditional fixed-accuracy transaction level models, the abstraction level of accuracy-adaptive models can be changed during simulation. This enables dynamic trade-off between simulation performance and accuracy. In this article we present a modeling style for systematic construction of accuracy-adaptive models. Additionally, we present the simulation semantics of those models in an abstract and rigorous manner which is independent of implementation language and simulation algorithm. This abstract formulation allows different sequential and parallel simulation schemes to be studied and compared. Based on the presented semantics, we present a high-performance parallel simulation scheme which offers considerable improvement over the performance of existing parallel simulation schemes.

Modeling and Simulation of Valve-controlled Cylinder System Based on Bond Graph

AMESim is a high performance modeling, simulation and dynamic analysis software. Based on the bond graph theory and the principle of hydraulic control, a simulation model of valve-controlled cylinder system in EPS fatigue testing bench is established by using AMESim software. By changing different factors of the hydraulic system and setting different target wave curves, the system effects with different conditions are discussed and the simulation results provide a theoretical guidance for modeling optimization.