parallel component technology improves the development efficiency and performance of parallel software. parallel component applications often need to be deployed on unstable heterogeneous cluster platforms. By analyzi...
详细信息
Vehicular fog computing constitutes an environment for execution of demanding computation and storage tasks. There formed a hierarchical decentralised and distributed architecture supports the resource-constrained dev...
详细信息
Stochastic block partitioning (SBP) is a community detection algorithm that is highly accurate even on graphs with a complex community structure, but its inherently serial nature hinders its widespread adoption by the...
详细信息
ISBN:
(纸本)9798350307924
Stochastic block partitioning (SBP) is a community detection algorithm that is highly accurate even on graphs with a complex community structure, but its inherently serial nature hinders its widespread adoption by the wider scientific community. To make it practical to analyze large real-world graphs with SBP, there is a growing need to parallelize and distribute the algorithm. The current state-of-the-art distributed SBP algorithm is a divide-and-conquer approach that limits communication between compute nodes until the end of inference. This leads to the breaking of computational dependencies, which causes convergence issues as the number of compute nodes increases and when the graph is sufficiently sparse. To address this shortcoming, we introduce EDiSt - an exact distributed stochastic block partitioning algorithm. Under EDiSt, compute nodes periodically share community assignments during inference. Due to this additional communication, EDiSt improves upon the divide-and-conquer algorithm by allowing it to scale out to a larger number of compute nodes without suffering from convergence issues, even on sparse graphs. We show that EDiSt provides speedups of up to 26.9x over the divide-and-conquer approach and speedups up to 44.0x over shared memory parallel SBP when scaled out to 64 compute nodes.
This research study investigates the impact of parallel programming techniques on the performance of searching and sorting algorithms. Traditional sequential algorithms have been the foundation of data processing for ...
详细信息
The influence of tactical evolution on agents' operational effectiveness in an air-to-ground attack (ATGA) mission is investigated. Firstly, the penetration-attack mission of 20-drones-swarm vs 10 air defense syst...
详细信息
The influence of tactical evolution on agents' operational effectiveness in an air-to-ground attack (ATGA) mission is investigated. Firstly, the penetration-attack mission of 20-drones-swarm vs 10 air defense systems is modeled and a distributed decision making model forATGAmission is constructed for the swarm. The decision making model contains several fuzzy inference systems(FISs), which consist of membership functions and inference rules, both presented in the form of "tactical genes". Then "tactical genes" are optimized by offline evolutionary computing, and a substantial enhancement of agents' operational effectiveness is witnessed in experimental results. Based on a high performance computing cluster and evolutionary computing framework, the idea that tactical genes can be optimized in an online manner is proved. Finally, experimental results show that shorter evolution period brings higher operational effectiveness.
Owing to population growth and industrialization, the need for electricity is at its apex, which creates stress on the grid due to the continuous consumption of power. To produce energy, natural resources are utilized...
详细信息
Graph processing has become increasingly popular and essential for solving complex problems in various domains, like social networks. However, the processing of graphs on a massive scale poses critical challenges, suc...
详细信息
ISBN:
(纸本)9783031488023;9783031488030
Graph processing has become increasingly popular and essential for solving complex problems in various domains, like social networks. However, the processing of graphs on a massive scale poses critical challenges, such as inefficient utilization of resources and energy. To bridge such challenges, the Graph-Massivizer project, funded by the Horizon Europe research and innovation program, conducts research and develops a high-performance, scalable, and sustainable platform for information processing and reasoning based on the massive graph (MG) representation of extreme data. This paper presents an initial architectural design for the Graph-Choreographer, one of the five Graph-Massivizer tools. We explain Graph-Choreographer's components and demonstrate how Graph-Choreographer can adopt the emerging serverless computing paradigm to process Basic Graph Operations (BGOs) as serverless functions across the computing continuum efficiently. We also present an early vision of our federated Function-as-a-Service (FaaS) testbed, which will be used to conduct experiments and assess the tool's performance.
In the era marked by rapid advancements in "green energy"and storage technologies, microgrids integrated with distributed generation principles emerge as a promising avenue for efficient power management. DC...
详细信息
The rapid adoption of Internet of Things (IoT) technologies and digital twins (DTs) is revolutionizing smart cities by facilitating improved real-time monitoring, simulation, and optimization of urban infrastructures....
详细信息
High-density EEG is a non-invasive measurement method with millisecond temporal resolution that allows us to monitor how the human brain operates under different conditions. The large amount of data combined with comp...
详细信息
ISBN:
(纸本)9783031488023;9783031488030
High-density EEG is a non-invasive measurement method with millisecond temporal resolution that allows us to monitor how the human brain operates under different conditions. The large amount of data combined with complex algorithms results in unmanageable execution times. Large-scaleGPU parallelism provides the means to drastically reduce the execution time of EEG analysis and bring the execution of large cohort studies (over thousand subjects) within reach. This paper describes our effort to implement various EEG algorithms for multiGPUpre-exascale supercomputers. Several challenges arise during thiswork, such as the high cost of data movement and synchronisation compared to computation. A performance-oriented end-to-end design approach is chosen to develop highlyscalable, GPU-only implementations of full processing pipelines and modules. Work related to the parallel design of the family of Empirical Mode Decomposition algorithms is described in detail with preliminary performance results of single-GPU implementations. The research will continue with multi-GPU algorithm design and implementation aiming to achieve scalability up to thousands of GPU cards.
暂无评论