Distributed Privacy-Aware Fast Selection Algorithm for Large-Scale Data

H, Liu; Chen, Jiming

doi:10.1109/tpds.2017.2761344

Cited by 7 publications

(1 citation statement)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The effective operation of these apps depends on this equilibrium. [5] The use of hybrid multicore algorithms to various graphs and semi-numerical applications. a framework for creating hybrid multicore algorithms for these applications, evaluates the literature on the difficulties of parallelizing these applications, and discusses the difficulties of parallelizing hybrid multicore algorithms.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Multicore Algorithms for Some Semi Numerical Applications and Graph

Sharma,

Tripathi,

Jain

2023

Recent Trends in Data Science and Its Applications

View full text Add to dashboard Cite

Due to its capacity to offer high-performance computing solutions for a variety of applications, hybrid multicore algorithms (HMAs) have grown in popularity in recent years. In this article, we offer a study on the use of HMAs to graphs and certain semi-numerical applications. We specifically look at how well HMAs perform in two different kinds of applications: graph algorithms and semi-numerical simulations. In terms of graph algorithms, we take into account a number of well-known issues, such as shortest path, minimal spanning tree, and graph clustering techniques. We put these algorithms into practise utilising both conventional CPU-based parallelization approaches and HMAs, and we evaluate how well they work with various graph sizes. We explore the challenge of employing partial differential equations to simulate the behaviour of complicated systems, such as fluid flow, for semi-numerical simulations (PDEs). We put into practise a hybrid strategy that combines GPU acceleration with CPU-based parallelization approaches, and we evaluate its performance against more conventional CPU-based parallelization strategies. showed both graph algorithms and semi-numerical simulations may significantly outperform conventional CPU-based parallelization strategies when using HMAs. In particular, HMAs can boost performance for graph algorithms up to a factor of two and for semi-numerical simulations up to a factor of five. Our findings show the potential of HMAs as a formidable tool for high-performance computing in graph algorithms and semi-numerical applications.

show abstract

Section: Introductionmentioning

confidence: 99%