Parallel Applications Research Articles

Monarch: A Durable Polymorphic Memory for Data Intensive Applications

3D die stacking has often been proposed to build large-scale DRAM-based caches. Unfortunately, the power and performance overheads of DRAM limit the efficiency of high-bandwidth memories. Also, DRAM is facing serious scalability challenges that make alternative technologies more appealing. This paper examines Monarch, a resistive 3D stacked memory based on a novel reconfigurable crosspoint array called XAM. The XAM array is capable of switching between random access and content-addressable modes, which enables Monarch (i) to better utilize the in-package bandwidth and (ii) to satisfy both the random access memory and associative search requirements of various applications. Moreover, the Monarch controller ensures a given target lifetime for the resistive stack. Our simulation results on a set of parallel memory-intensive applications indicate that Monarch outperforms an ideal DRAM caching by <inline-formula><tex-math notation="LaTeX">$1.21\times$</tex-math></inline-formula> on average. For in-memory hash table and string matching workloads, Monarch improves performance up to <inline-formula><tex-math notation="LaTeX">$12\times$</tex-math></inline-formula> over the conventional high bandwidth memories.

A Compression Router for Low-Latency Network-on-Chip

Network-on-Chips (NoCs) are important components for scalable many-core processors. Because the performance of parallel applications is usually sensitive to the latency of NoCs, reducing it is a primary requirement. In this study, a compression router that hides the (de)compression-operation delay is proposed. The compression router (de)compresses the contents of the incoming packet before the switch arbitration is completed, thus shortening the packet length without latency penalty and reducing the network injection-and-ejection latency. Evaluation results show that the compression router improves up to 33% of the parallel application performance (conjugate gradients (CG), fast Fourier transform (FT), integer sort (IS), and traveling salesman problem (TSP)) and 63% of the effective network throughput by 1.8 compression ratio on NoC. The cost is an increase in router area and its energy consumption by 0.22mm2 and 1.6 times compared to the conventional virtual-channel router. Another finding is that off-loading the decompressor onto a network interface decreases the compression-router area by 57% at the expense of the moderate increase in communication latency.

The Role of Idle Waves, Desynchronization, and Bottleneck Evasion in the Performance of Parallel Programs

The performance of highly parallel applications on distributed-memory systems is influenced by many factors. Analytic performance modeling techniques aim to provide insight into performance limitations and are often the starting point of optimization efforts. However, coupling analytic models across the system hierarchy (socket, node, network) fails to encompass the intricate interplay between the program code and the hardware, especially when execution and communication bottlenecks are involved. In this paper we investigate the effect of "bottleneck evasion" and how it can lead to automatic overlap of communication overhead with computation. Bottleneck evasion leads to a gradual loss of the initial bulk-synchronous behavior of a parallel code so that its processes become desynchronized. This occurs most prominently in memory-bound programs, which is why we choose memory-bound benchmark and application codes, specifically an MPI-augmented STREAM Triad, sparse matrix-vector multiplication, and a collective-avoiding Chebyshev filter diagonalization code to demonstrate the consequences of desynchronization on two different supercomputing platforms. We investigate the role of idle waves as possible triggers for desynchronization and show the impact of automatic asynchronous communication for a spectrum of code properties and parameters, such as saturation point, matrix structures, domain decomposition, and communication concurrency. Our findings reveal how eliminating synchronization points (such as collective communication or barriers) precipitates performance improvements that go beyond what can be expected by simply subtracting the overhead of the collective from the overall runtime.

RECOGNITION OF A LANDSCAPE TECHNO-SPHERE AS A NEW GEOSPHERE

The second half of XX century – the beginning of XXI century are characterized by the active development of a global ecological crisis, which is described by a critical state of the environment and excessive anthropogenization of a geographical sphere. The changes which lead to the formation of absolutely new techno-genic landscapes occur at very fast rates in current geo-spheres. The purpose of the work is to substantiate the recognition of a new geo-sphere – a landscape techno- sphere – based on the previous experience and our field research. The main tasks of the research are to analyze specific features and properties of a landscape techno-sphere; to identify its upper and lower borders; to study a structural organization of all options of a landscape sphere with available techno-substances; to characterize briefly the main stages of the development of a techno- sphere.&#x0D; The research of a landscape techno-sphere is a complicated process and it is based on the use of classical and innovative methodological principles of contemporary geography. A parallel application of three scientific paradigms is the foundation: system, model and ecological. They do not contradict each other; they rather compliment the studying of the interaction between nature and engineering. A system approach allows considering a techno-sphere as a complicated system of a lower taxonomic range of a landscape sphere and a geographical sphere. The following generally accepted scientific methods are used in the paper: analysis, synthesis, comparison, generalization, systematization, induction and deductions.&#x0D; The article substantiates the existence of a new geosphere – the landscape techno-sphere, which is formed by the planetary set of landscape-technical systems. it is noted that it does not have a continuous distribution, but is constantly increasing in size and gradually goes beyond the landscape. The specific features and properties of the landscape techno-sphere are analyzed in detail. Taking into account the criterion of the presence of a zone of direct contract of three blocks of landscape-technical systems (managerial, technical and natural), the upper and lower limits of the landscape techno-sphere are identified. The structure of all variants of the landscape sphere in the presence of technological substances is considered. The main periods of development of the landscape techno-sphere are briefly described. It is concluded that man will never be able to control landscape-technical systems on a planetary scale. However, applying the methods and principles of engineering landscape research, it is possible to achieve optimal interaction of the three blocks of landscape-technical systems and to extend their functional suitability.

An efficient fault tolerance scheme based enhanced firefly optimization for virtual machine placement in cloud computing

SummaryThe virtual machine placement for the highly reliable cloud application is considered as one of the challenging and critical issues. To tackle such an issue, this article proposes the enhanced firefly algorithm based virtual machine placement model. But the migration time of the virtual machine placement is high and to reduce the migration time of the virtual machine placement, this article utilizes the K‐means clustering algorithm. In addition, to obtain the optimal cluster for the virtual machine placement, the adaptive particle swarm optimization with the coyote optimization algorithm is employed. The experimental results are conducted for the proposed approach using various measures such as transmission overhead, total execution time, packet size, parallel applications numbers, and virtual machine numbers. The results demonstrate that the proposed method offers improved performance and an optimal virtual machine placement scheme with respect to the various constraint factors. The evaluation exposes that the proposed method offers less execution time when compared to other methods.

Dynamic Analysis and Design of an SMCNN of a 6-DOF Parallel Mechanisms using Translation Driver Motor

—Recently, many technological improvement apply in the discovery of various designs of haptic devices. Several mechanism structures including serial, parallel, and hybrid-kinematic manipulators can be considered for making a haptics device. The most successful haptic mechanisms are parallel-type, because of low moving inertia, large force reflection, and high stiffness. This research shows the 6-DOF parallel haptics device based on the parallel mechanism using a translation driver motor mounted on each leg. Firstly, we introduce a 6-DOF parallel mechanism using a translation driver motor haptics device model. Due to the unsure parameters, we focus on solving the mathematics model with the nonlinear parameters of the 6-DOF parallel mechanism. Secondly, to fix the kinematics and dynamics nonlinear uncertainties parameters, the SMCNN controller for 6-DOF parallel mechanism application using a translation driver motor is designed. The Sliding model control base on artificial intelligence neural network is used to calculate the unsure factors. In this technique, to prove the stability of the system the Lyapunop theory is used. Finally, the authors the simulation results of two control algorithms with different uncertain components are presented and comparing them to demonstrate the effectiveness of the new control method. The control method is demonstrated by way of implementing the set of rules in artificial surroundings with realistic parameters, in which the received consequences are fairly promising. The obtained from SMCNN algorithm results are highly promising and accurate.

Accelerating Parallel Applications Based on Graph Reordering for Random Network Topologies

The Message Passing Interface (MPI) is a crucial programming tool for enabling communication between processes in parallel applications. The goal of MPI users is to allocate tasks to processors in a way that maximizes both spatial and temporal locality in the network. However, this can be challenging, especially in large-scale networks where maximizing processor locality may not be feasible at runtime. To address this issue, we propose the use of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Hamorder</i> , an offline node reassignment approach that takes into account physical processor locations based on graph reordering for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Random</i> network topologies. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Hamorder</i> aims to optimize task mapping for improved performance in parallel applications, whether for multiple tasks or within a single task. Additionally, we investigate the potential of improving MPI applications through runtime parameter tuning based on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Hamorder</i> . Our evaluation results show that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Hamorder</i> provides a 27.3% improvement in performance compared to the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Gorder</i> algorithm on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Random</i> topologies, which is a state-of-the-art solution designed with the aim of enhancing cache locality and achieves this goal by rearranging the vertices of a graph in a way that places the vertices that are typically accessed together in close proximity. Moreover, our autotuning framework using <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Hamorder</i> results in an average speedup of 1.38x for targeted MPI applications by searching through various runtime parameter combinations.

Integrating parallelism and asynchrony for high-performance software development

This article delves into the crucial roles of parallelism and asynchrony in the development of high-performance software programs. It provides an insightful exploration into how these methodologies enhance computing systems' efficiency and performance. The discussion begins with an overview of high-performance programs and their importance in modern computing, followed by a detailed analysis of parallelism, including its hardware and software aspects, various programming models, and diverse applications in fields like computational tasks, data processing, and machine learning. The article then shifts focus to asynchrony, elaborating on its implementation through callback functions, promises, and the async/await syntax. It highlights the significant benefits of asynchrony in improving program responsiveness and efficiency, especially in I/O-bound and network-intensive operations. A critical section of the article examines the combined application of parallelism and asynchrony in different domains, notably in Graphics Processing Unit (GPU) computations, distributed systems, and web development. This combination is shown to be particularly effective in handling Big Data, where rapid and efficient data processing is essential. The article concludes by emphasizing the indispensable nature of these approaches in modern software development. Parallelism and asynchrony are not merely performance enhancement tools but are fundamental in building innovative, resilient, and scalable software solutions. This comprehensive exploration underscores the ongoing relevance of these methodologies and their potential to drive further advancements in the field of high-performance computing.

Топологічні аспекти моделювання паралельних обчислень при розв’язанні динамічних задач на базі інтерфейсу передачі повідомлень

The article deals with the development, substantiation and research of analytical models of data transmission when solving complex multidimensional dynamic problems. Models of interprocessor exchange are obtained for computing systems with distributed memory and topological architectures: ring, 2D-torus, hypercube, etc. The application of double arithmetic based on the GMP-library and its effect on the temporal characteristics of parallel applications are considered. The dependence of the parallelism characteristics of numerical algorithms when using the streaming SIMD-extension is analyzed. The modern stage of development of computer and information technologies has one of its directions - the use of the idea of parallelism in order to reduce the time of solving multidimensional problems that have a high temporal complexity or are even NP-complete. Despite the significant increase in the productivity of existing parallel systems, reducing the cost of components for their construction, development and implementation of parallel methods remains the most difficult problem due to the lack of study of the internal structure of algorithms and their properties. Thus, the introduction of parallel computing requires not so much the mastering of parallel data processing as the development of parallel algorithms based on the parallelization of sequential ones or the construction of fundamentally new, more efficient parallel methods. That is why the analysis of existing parallel algorithms, studying the possibilities of improving their characteristics is an important, promising and practically demanded task. The purpose of the work is to increase the efficiency of solving complex multidimensional dynamic problems implemented on parallel architectures of distributed memory using the message passing interface (MPI) due to the reduction of communication time. Tasks of the work: firstly, it is the development of analytical models of multiple data transfer operations for different topologies of connection of processors, substantiation of the models and research of the quality of parallel calculations of various classes of numerical methods for cluster systems. The possibility and effectiveness of using double arithmetic (GMP-library) and its influence on the temporal characteristics of parallel implementation are separately investigated. Additionally, the paper analyzes the dependence of the parallelism characteristics of the methods on the use of Streaming SIMD-Extension (SSE). The scientific novelty of the field lies in the development and development of analytical models for processes of interprocessor exchange given by cluster systems, which allow to increase the efficiency of parallel implementations of numerical methods in rich dynamic tasks for a short period of time on a communication warehouse. The practical significance of the field is in the development of split models for the analysis of the efficiency of collective operations of multiple data transfer within the MPI interface, which is the current standard for software implementation of robots with shared memory. Possibility of arranging floating arithmetic based on the GMP library was added, which is especially important for the development of folding problems of dynamic problems with singularities.

RouteReplies: Alleviating Long Latency in Many-Chip-Module GPUs

GPU chip module count is expected to keep increasing to meet the strong scaling demands of parallel applications. In many-chip-module GPUs, memory access latency seriously limits the performance since the transferring latency between different GPU modules is very high, which cannot be easily hidden by switching between different ready threads. To handle this problem, we propose RouteReplies, which enables a GPU module to fetch data from other GPU modules in the routing path. Leveraging the data locality between different GPU modules, RouteReplies significantly reduces the memory access latency since the memory request does not need to fetch data from the faraway memory partition. For a set of applications exhibiting varying degrees of inter-module locality, RouteReplies reduces memory access latency and increases performance by 54.8% on average (up to 364.8%).

On the Parallel Application of Statutory Compensation and Punitive Compensation in Trademark Infringement

On the Parallel Application of Statutory Compensation and Punitive Compensation in Trademark Infringement

Optimizing Iterative Data-flow Scientific Applications using Directed Cyclic Graphs

Data-flow programming models have become a popular choice for writing parallel applications as an alternative to traditional work-sharing parallelism. They are better suited to write applications with irregular parallelism that can present load imbalance. However, these programming models suffer from overheads related to task creation, scheduling and dependency management, limiting performance and scalability when tasks become too small. At the same time, many HPC applications implement iterative methods or multi-step simulations that create the same directed acyclic graphs of tasks on each iteration. By giving application programmers a way to express that a specific loop is creating the same task pattern on each iteration, we can create a single task DAG once and transform it into a cyclic graph. This cyclic graph is then reused for successive iterations, minimizing task creation and dependency management overhead. This paper presents the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">taskiter</i> , a new construct we propose for the OmpSs-2 and OpenMP programming models, allowing the use of directed cyclic task graphs (DCTG) to minimize runtime overheads. Moreover, we present a simple <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">immediate successor</i> locality-aware heuristic that minimizes task scheduling overhead by bypassing the runtime task scheduler. We evaluate the implementation of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">taskiter</i> and the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">immediate successor</i> heuristic in 8 iterative benchmarks. Using small task granularities, we obtain a geometric mean speedup of 2.56x over the reference OmpSs-2 implementation, and a 3.77x and 5.2x speedup over the LLVM and GCC OpenMP runtimes, respectively.

Theoretical and methodological aspects of the essence of the concept of “Public governance”

The article highlights the main approaches to defining the main theoretical and methodological aspects of the essence of the concept of “public administration”; identified scientific approaches to the definitions of public administration; outlines a scientific discussion around the theoretical foundations of the process of public administration. The change of scientific approaches to understanding the essence of the concept of “public administration” and the parallel application of the concept of “public administration” in the context of the development of democracy and civil society in Ukraine. Emphasis is placed on the need to increase the efficiency of public administration at all levels and to revise the model of public administration, which should aim to meet the demands of the population throughout the country. It is emphasized that management is a complex social process that involves purposeful influence on human activity, society in order to obtain a specific expected result. It is stated that public administration is a kind of socially useful activity carried out by a certain set of entities, including public authorities, which are interconnected with public administration, because public administration is carried out as part of society itself. Public administration arises as a result of a certain social agreement between the state and society. The subject and at the same time the object of public administration is civil society, public associations, organizations, public authorities. Emphasis is placed on the fact that the state policy of decentralization involves the transformation at the territorial level of local government into a real entity with powers and resources that will address all organizational, social, financial and other issues in a given area. Public administration is a complex process of state and municipal administration, the purpose of which is to control the activities of public authorities and local self-government by citizens through the system of suffrage. Public administration is a type of socially useful activity carried out by a certain set of entities, in particular public authorities, which are interconnected with public administration, because public administration is carried out as part of society itself. Public administration arises as a result of a certain social agreement between the state and society. The subject and at the same time the object of public administration is civil society, public associations, organizations, public authorities. Public administration is a system consisting of organizational, political, legal, economic mechanisms of public administration and the mechanism of local self-government. The main idea of public administration is the management of society together with society itself, where partners of public authorities in addressing a wide range of socially important and important issues are public associations, religious organizations, various civil society institutions, commercial institutions, etc., legal entities and individuals. exercising public control over public administration decisions, etc.

Jdebug: A Fast, Non-Intrusive and Scalable Fault Locating Tool for Ten-Million-Scale Parallel Applications

This article presents Jdebug, a fast, non-intrusive and scalable fault locating tool for extreme-scale parallel applications. Large-scale debugging has drawn more attention with the increasing scale of supercomputers and applications. To eliminate program intrusion caused by traditional instrumentation or interception during debugging information acquisition, we introduce the out-of-band management into large-scale debugging. We propose a rapid information gathering scheme that separates user and debugging traffic to solve scalability problem and to eliminate program interference during merging data. Observations of Program Counters (PC) and performance characteristics in suspended applications find abnormalities and help locate abnormal threads caused by software errors or hardware failures effectively. Evaluation shows that Jdebug collects PCs of over 20 million cores on the new Sunway supercomputer within 1.97 seconds, and can locate the abnormal threads in 1.4 seconds with an accuracy of 92.5%. In the running test of three fundamental benchmarks (HPL, HPCG, Graph500) and seventeen real-world applications, Jdebug quickly and accurately locates abnormal threads to help find scalability errors and hardware failures including memory access failures, communication failures, and execution component failures, which validates its effectiveness.

Leveraging Code Snippets to Detect Variations in the Performance of HPC Systems

Variations in the performance of parallel and distributed systems are becoming increasingly challenging. The runtimes of different executions can vary greatly even with a fixed number of computing nodes. Many HPC applications on supercomputers exhibit such variance. This not only leads to unpredictable execution times, but also renders the system’s behavior unintuitive. The efficient online detection of variations in performance is an open problem in HPC research. To solve it, we propose an approach, called <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vSensor</small> , to detect variations in the performance of systems. The key finding of this study is that the source code of programs can better represent performance at runtime than an external detector. Specifically, many HPC applications contain code snippets that are fixed workload patterns of execution, e.g., the workload of an invariant quantity and a linearly growing workload. This observation allows us to automatically identify these snippets of workload-related code and use them to detect variations in performance. We evaluate <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vSensor</small> on the Tianhe-2A system with a large number of parallel applications, and the results indicate that it can efficiently identify variations in system performance. The average overhead of 4,096 processes is less than 6% for fixed-workload v-sensors. We identify a problematic node with slow memory by using <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vSensor</small> that degrades the performance of the program by 21%. A serious issue with network performance is also detected that slows down the Tianhe-2A system by 3.37 times for an HPC kernel.

Detecting Performance Variance for Parallel Applications Without Source Code

For parallel applications, performance variance is a critical issue that can degrade performance and make applications’ behavior difficult to explain. Therefore, users and application developers should be able to detect and diagnose performance variance. Previous detection methods either introduce too much overhead and slow down applications, or rely on nontrivial source code analysis, which is impractical for production-run parallel systems. In this article, we propose <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vapro</small> , a framework for detecting and diagnosing performance variance in production-run parallel systems. Our method is based on an observation that most parallel programs contain code snippets that are executed repeatedly with a fixed workload and can be utilized to detect performance variance. We present State Transition Graph (STG) to track program execution and then do light-weight workload analysis on STG to locate performance variance. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vapro</small> is able to successfully identify these snippets at runtime even without program source code. To diagnose the discovered variation, <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vapro</small> uses a progressive diagnosis method based on a hybrid model combining variance breakdown and statistical analysis. According to evaluating results, <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vapro</small> 's performance overhead is only 1.38% on average. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vapro</small> can identify performance variance in real applications caused by hardware issues, such as memory and IO. The standard deviation of the execution time is decreased by up to 73.5% when the identified variance is fixed. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vapro</small> achieves 30.0% larger detection coverage than the state-of-the-art variance detection approach based on source code analysis.

Populism and authoritarianism in the EU: what is the role of the euro?

ABSTRACT The article examines the impact of the euro on the political system through contrasting the experiences of eurozone countries with euro outsiders using a historical institutionalist perspective. While following the global financial crisis, many have argued that the homogeneity of rules within the Economic and Monetary Union (EMU) facilitate the rise of populist movements, authoritarian challengers have emerged only outside the eurozone. The parallel application of Rodrik’s political economy trilemma and the ordoliberal tradition of the Freiburg school to the EMU provides the theoretical background for explaining this puzzle. A most- similar research design based on the comparative case study of Greece and Hungary is used to uncover the causal processes through which the EMU contributes to the resilience of liberal democracy. It is shown that while the limits on economic policy autonomy strongly hinder authoritarian tendencies within the eurozone even under a populist government, being outside the EMU is a necessary though not sufficient condition for the entrenchment of authoritarian rule.

Near-optimal energy-efficient partial-duplication task mapping of real-time parallel applications

Minimizing energy consumption, as well as meeting real-time and reliability constraints, are major goals during system deployment. When complex platforms, such as multicore architectures with DVFS, and parallel applications are considered, these goals are significantly impacted by task mapping. To minimize energy consumption, while meeting real-time and reliability constraints, this work proposes a task mapping approach to jointly solve the problem of task allocation, task scheduling, frequency assignment, and task duplication. A novel heuristic algorithm is proposed to cope with this NP-hard problem, consisting of a pruning phase, which maintains only the task configurations that satisfy reliability constraints, and a mapping phase, which minimizes total energy consumption under real-time and precedence constraints. The obtained results show that the proposed heuristic obtains near-optimal results, with low computation time, compared to optimal solvers, while it achieves better energy consumption and finds slightly more solutions compared to other heuristic approaches.

Performance prediction of parallel applications using artificial neuronal network and graph representation

SummaryModeling and predicting the performance of HPC applications is exceedingly important for several purposes. These incorporate scheduling and managing tasks, understanding application behavior, and estimating resource requirements. Many issues impact the performance of parallel applications, including data decomposition, load balancing, and inter processors communication. Predicting the performance of parallelism accurately is a challenging task, especially since it depends on several hardware and software parameters that are difficult to identify. The graph representation is an efficient way to model a parallel application. Based on this representation, this study purports to predict the optimal number of partitions (processors) to run efficiently a parallel application on multi‐core architectures using an artificial neural network. The numerical simulation applications are modeled as a graph of tasks and their associated data. Using the characteristics of the graphs as inputs of the ANN model, we obtained satisfactory results with an accuracy of 91% for the optimal number of processors and 93% for the maximum speedup.

Japanese Pharmaceutical Regulations of Engineered Viral Vectors for Medical Use Compared With Those in the United States and the European Union.

Japanese Pharmaceutical Regulations of Engineered Viral Vectors for Medical Use Compared With Those in the United States and the European Union.