repository
대규모 그래프 처리를 위한 Label-Propagation 기반 병렬 그래프 파티션 기법
- Alternative Title
- Minho Bae
- Author(s)
- 배민호
- Alternative Author(s)
- Minho Bae
- Advisor
- 오상윤
- Department
- 일반대학원 컴퓨터공학과
- Publisher
- The Graduate School, Ajou University
- Publication Year
- 2019-08
- Language
- eng
- Alternative Abstract
- The increasing importance of graph data in various fields makes the efficient processing of large-scale graph data highly necessary where well-balanced graph partitioning is a vital component of parallel/distributed graph processing. The goal of graph partitioning is to obtain a well-balanced graph topology that balances the size of each partition while reducing the number of edge-cuts. Moreover, the graph partitioning algorithm should achieve high performance and scalability. In this paper, we present a novel graph-partitioning algorithm that ensures low edge-cuts and high-performance processing capability for parallel processing. Based on the label propagation algorithm, we propose the formulas to improve the degree of edge-cuts and to achieve fast convergence. By removing the necessity of processing the label propagation for all nodes, our approach processes the label propagation of candidate nodes based on a proposed score metric. Our proposed algorithm introduces a stabilization phase in which remote and highly connected nodes are relocated to avoid the algorithm becoming trapped around local optima. Comparison results show that a prototype based on the proposed algorithm outperforms other well-known parallel graph-partitioning frameworks in terms of speed and balance.
- Fulltext
- Appears in Collections:
- Graduate School of Ajou University > Department of Computer Engineering > 4. Theses(Ph.D)
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
-
- License
- STATISTICS
- Total Visit :2,960,078
- Total Download :1,694
- Today View :6,464
