Search

Summary: Task‐based systems have become popular due to their ability to utilize the computational power of complex heterogeneous systems. A typical programming model used is the Sequential Task Flow (STF) model, which unfortunately only supports static task graphs. This can result in submission overhead and a static task graph that is not well‐suited for execution on heterogeneous systems. A common approach is to find a balance between the granularity needed for accelerator devices and the granularity required by CPU cores to achieve optimal performance. To address these issues, we have extended the STF model in the StarPU runtime system by introducing the concept of hierarchical tasks. This allows for a more dynamic task graph and, when combined with an automatic data manager, it is possible to adjust granularity at runtime to best match the targeted computing resource. That data manager makes it possible to switch between various data layout without programmer input and allows us to enforce the correctness of the DAG as hierarchical tasks alter it during runtime. Additionally, submission overhead is reduced by using large‐grain hierarchical tasks, as the submission process can now be done in parallel. We have shown that the hierarchical task model is correct and have conducted an early evaluation on shared memory heterogeneous systems using the Chameleon dense linear algebra library. [ABSTRACT FROM AUTHOR]

Certain workloads such as in‐memory databases are inherently hard to scale‐out and rely on cache‐coherent scale‐up non‐uniform memory access (NUMA) systems to keep up with the ever‐increasing demand for compute resources. However, many parallel programming frameworks such as OpenMP do not make efficient use of large scale‐up NUMA systems as they do not consider data locality sufficiently. In this work, we present PGASUS, a C++ framework for NUMA‐aware application development that provides integrated facilities for NUMA‐aware task parallelism and data placement. The framework is based on an extensive review of parallel programming languages and frameworks to incorporate the best practices of the field. In a comprehensive evaluation, we demonstrate that PGASUS provides average performance improvements of 1.56× and peak performance improvements of up to 4.67×across a wide range of workloads. [ABSTRACT FROM AUTHOR]

Dual hesitant fuzzy elements (DHFEs) are suitable to express hesitant possible preferred and nonpreferred judgments of decision makers. Preference relation is an important tool in decision making that only needs the decision makers to compare a pair of objects at one time. This study focuses on decision making with dual hesitant fuzzy preference relations (DHFPRs). Considering the consistency, an additive consistency concept is defined. Meanwhile, the property of the new concept is studied. Using this consistency concept, a method for assessing the additive consistency of DHFPRs is offered. To extend the application of DHFPRs, a programming model to determine the missing DHFEs in incomplete DHFPRs is built, which have the highest additive consistency level for the known ones. Two equivalent methods to calculate the priority vector are offered. One method obtains the probabilistic dual hesitant fuzzy priority vector, and the other derives the intuitionistic fuzzy priority vector. Furthermore, a consensus index is defined to measure the consensus of individual opinions in group decision making (GDM), and an interactive method for increasing the consensus level is offered. On the basis of the additive consistency and consensus, an algorithm to GDM with DHFPRs is offered that can address inconsistent and incomplete cases. Finally, a practical example about evaluating color TV is provided to demonstrate the usefulness of the new procedure. [ABSTRACT FROM AUTHOR]

For commuters, the workplace is the second most important place to be able to charge electric vehicles after their own residence. In order to satisfy the charging demand near to the workplaces of commuters, a location and capacity model for charging stations is set up with the aim of minimizing the overall cost, composed of the charging station construction cost and the generalized cost of users. The charging demand of commuting electric vehicles is divided into rigid demand and elastic demand, which depends on whether the electric vehicle has enough power to complete the rest of its daily travel requirements. By introducing a user convenience coefficient to quantify the user's charging expectation, the number of electric vehicles to be served and the demand for charging at the stations are obtained. Further, the optimal sites for charging stations and the quantities of different levels of charging facility equipped are determined by the location and capacity model. An example shows that, as the threshold of the convenience coefficient decreases, the demand for electric energy increases gradually, and the construction becomes less economical. Meanwhile, the location of the charging station is insensitive to the change of electric energy demand. The threshold value for the convenience coefficient is put forward for different planning purposes, providing a reference for planning site selection and quantity of chargers under elastic demand. © 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

Intuitionistic fuzzy multiplicative preference relations (IFMPRs), as an extension of multiplicative preference relations, can denote the decision‐makers' (DMs') preferred and nonpreferred degrees simultaneously. Just as any other type of preference relations, consistency is crucial to guarantee the rational ranking orders. Thus, this paper introduces a new consistent concept for IFMPRs that is a natural extension of crisp case and overcomes the issues in the previous concepts of consistency. To judge the consistency of IFMPRs, several programming models are constructed, and an approach to deriving completely consistent IFMPRs is presented. Considering incomplete case, consistency‐based models are built to determine missing values that can address incomplete IFMPRs with the ignored objects, namely, all information for them is unknown. After that, group decision‐making with IFMPRs is studied. To measure the agreement degree between the DMs' individual IFMPRs, a new consensus index is defined, and an interactive algorithm to improve the consensus is offered. Based on the consistency and consensus analysis, a new method to group decision‐making with IFMPRs is developed. Finally, case studies are offered to show the application of the new procedure and to compare it with previous methods. [ABSTRACT FROM AUTHOR]

To address the preferred and nonpreferred degrees of linguistic variables, this paper introduces intuitionistic linguistic preference relations (ILPRs) that apply intuitionistic linguistic variables (ILVs) to denote the decision makers' preferences. To judge the consistency of ILPRs, a consistent concept is introduced, and a consistency index is defined. When ILPRs are unacceptably inconsistent, a method to improve the consistency is introduced. Then, an approach to rank ILVs is introduced. In some situations where ILPRs might be incomplete, the consistency‐based linear programming model is constructed to evaluate the missing values. Considering group decision making, a group consensus index is defined, and its several desirable properties are discussed. Meanwhile, an acceptability‐based consistency and consensus approach is developed, and the associated example is offered to show the efficiency of the proposed procedure. [ABSTRACT FROM AUTHOR]

Hesitant information is powerful and flexible to denote decision maker's judgments. Hesitant multiplicative preference relations (HMPRs) own the advantages of preference relations and hesitant fuzzy sets that permit the decision makers (DMs) to compare objects by using several values. Just as other types of preference relations, how to derive the priority weight vector is a crucial step. According to the principle of the consistency concept for multiplicative preference relations, this paper first introduces a new consistency concept for HMPRs, which avoids the disadvantages of the previous ones. Using the new concept, models to judge the consistency of HMPRs are built. Then, a consistency probability‐based method to derive the hesitant fuzzy priority weight vector from HMPRs is offered. Considering the incomplete case, consistency‐based programming models to determine the missing values are constructed. To address group decision making with HMPRs, a distance measure is defined to determine the weights of the DMs, and a consensus index is proposed. Then, a consistency and consensus‐based group decision‐making algorithm is performed. Finally, two practical examples, an investment problem and a water conservancy problem are offered to illustrate the feasibility and efficiency of the new algorithm. Comparison analysis from the numerical and theoretical aspects verifies the potential application of the new procedure. [ABSTRACT FROM AUTHOR]

The Internet of things (IoT) has attracted a great deal of research and industry attention recently and is envisaged to support diverse emerging domains including smart cities, health informatics, and smart sensory platforms. Operating system (OS) support for IoT plays a pivotal role in developing scalable and interoperable applications that are reliable and efficient. IoT is implemented by both high‐end and low‐end devices that require OSs. Recently, the authors have witnessed a diversity of OSs emerging into the IoT environment to facilitate IoT deployments and developments. In this study, they present a comprehensive overview of the common and existing open‐source OSs for IoT. Each OS is described in detail based on a set of designing and developmental aspects that they established. These aspects include architecture and kernel, programming model, scheduling, memory management, networking protocols support, simulator support, security, power consumption, and support for multimedia. They present a taxonomy of the current IoT open‐source OSs. The objective of this survey is to provide a well‐structured guide to developers and researchers to determine the most appropriate OS for each specific IoT devices/applications based on their functional and non‐functional requirements. They remark that this is the first such tutorial style paper on IoT OSs. [ABSTRACT FROM AUTHOR]

Abstract: Norway maintains a complex system of activity or type specific coupled payments which account for a large share of farm income. Most of the payment rates are negatively related to farm size and are higher in remote areas compared to central regions. We present and use a newly developed recursive‐dynamic multi‐commodity model (Agrispace) with CES production functions depicting regional farm clusters derived from the full farm population. Using this model, we simulate impacts of current and alternative subsidy policies on production, prices, input use, income and farm structural change. Mapping cluster results to each farm along with behavioural rules allows estimation of individual profits and farm exits. Our results indicate that, in the short run, the current policy regime seems to support the policy objective of maintaining a variety of farms in all parts of Norway. In the long run, farm structural change is less affected by a policy reform that leaves total support levels unchanged. [ABSTRACT FROM AUTHOR]

Nowadays, computing hardware continues to move toward more parallelism and more heterogeneity, to obtain more computing power. From personal computers to supercomputers, we can find several levels of parallelism expressed by the interconnections of multi-core and many-core accelerators. On the other hand, computing software needs to adapt to this trend, and programmers can use parallel programming models (PPM) to fulfil this difficult task. There are different PPMs available that are based on tasks, directives, or low-level languages or library. These offer higher or lower abstraction levels from the architecture by handling their own syntax. However, to offer an efficient PPM with a greater (additional) high-level abstraction level while saving on performance, one idea is to restrict this to a specific domain and to adapt it to a family of applications. In the present study, we propose a high-level PPM specific to digital signal-processing applications. It is based on data-flow graph models of computation, and a dynamic run-time model of execution (StarPU). We show how the user can easily express this digital signal-processing application and can take advantage of task, data, and graph parallelism in the implementation, to enhance the performances of targeted heterogeneous clusters composed of CPUs and different accelerators (e.g., GPU and Xeon Phi). Copyright © 2015 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

The rapid rise of OpenMP as the preferred parallel programming paradigm for small-to-medium scale parallelism could slow unless OpenMP can show capabilities for becoming the model-of-choice for large scale high-performance parallel computing in the coming decade. The main stumbling block for the adaptation of OpenMP to distributed shared memory (DSM) machines, which are based on architectures like cc-NUMA, stems from the lack of capabilities for data placement among processors and threads for achieving data locality. The absence of such a mechanism causes remote memory accesses and inefficient cache memory use, both of which lead to poor performance. This paper presents a simple software programming approach called copy-inside–copy-back (CC) that exploits the data privatization mechanism of OpenMP for data placement and replacement. This technique enables one to distribute data manually without taking away control and flexibility from the programmer and is thus an alternative to the automat and implicit approaches. Moreover, the CC approach improves on the OpenMP-SPMD style of programming that makes the development process of an OpenMP application more structured and simpler. The CC technique was tested and analyzed using the NAS Parallel Benchmarks on SGI Origin 2000 multiprocessor machines. This study shows that OpenMP improves performance of coarse-grained parallelism, although a fast copy mechanism is essential. Copyright © 2004 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

This paper presents a mathematical model for all matter-energy processing subsystems at the level of the society, specifically India. It explores India's choices in the food and energy sectors over the coming decades. Alternative land intensive, irrigation energy intensive, and fertilizer intensive techniques of food production are identified using a nonlinear programming model. The land saved is devoted to growing firewood. The optimum combination of railway (steam, diesel, and electric traction) and road (automobiles, diesel trucks, and diesel and gasoline buses) transport is determined. For the oil sector, two alternative sources of supply of crude oil and petroleum products are included, namely, domestic production and imports. The optimum choice is determined through a linear programming model. While the model is basically a static one, designed to determine the optimal choice for the target year of 2000-2001, certain intertemporal detail is incorporated for electricity generation. The model minimizes the costs of meeting the needs for food, transport in terms of passenger kilometers and goods per ton per kilometer, energy needs for domestic cooking and lighting, and the energy needs of the rest of the economy. [ABSTRACT FROM AUTHOR]

Summary: The rise of heterogeneous systems has given place to great challenges for users as they involve new concepts, restrictions, and frameworks. Their exploitation is further complicated in the context of distributed memory systems, which require the usage of additional different programming paradigms and tools. In this paper, we propose a novel approach to program heterogeneous clusters that is based on high‐level abstractions such as tiles and hierarchical decomposition combined with the powerful APIs that data types and embedded languages can provide in languages such as C++. Rather than building our proposal from scratch, we have implemented it as a natural integration of the existing Hierarchically Tiled Arrays (HTA) and Heterogeneous Programming Library (HPL) projects, ie, the first one being focused on distributed computing and the second one on heterogeneous processing. The result, called Heterogeneous Hierarchically Tiled Arrays (H2TA), is very intuitive and easy to use thanks to the global view of the data and the single‐threaded view of the execution that it provides at cluster level together with the transparency it provides with respect to the management of the heterogeneous devices. An evaluation comparing our proposal with MPI‐based implementations shows its large programmability advantages and the reasonable overhead incurred. [ABSTRACT FROM AUTHOR]