DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Min-Soo | ko |
dc.contributor.author | Lee, Sangyeon | ko |
dc.contributor.author | Han, Wook-Shin | ko |
dc.contributor.author | Park, Himchan | ko |
dc.contributor.author | Lee, Jeong-Hoon | ko |
dc.date.accessioned | 2020-06-02T01:21:21Z | - |
dc.date.available | 2020-06-02T01:21:21Z | - |
dc.date.created | 2020-05-22 | - |
dc.date.created | 2020-05-22 | - |
dc.date.created | 2020-05-22 | - |
dc.date.issued | 2015-04 | - |
dc.identifier.citation | IEEE International Conference on Data Engineering ICDE, pp.2658 - 2671 | - |
dc.identifier.uri | http://hdl.handle.net/10203/274440 | - |
dc.description.abstract | Computing connected components is a core operation on graph data. Since billion-scale graphs cannot be resident in memory of a single server, several approaches based on distributed machines have recently been proposed. The representative methods are Hash-To-Min and PowerGraph. Hash-To-Min is the state-of-the art disk-based distributed method which minimizes the number of MapReduce rounds. PowerGraph is the-state-of-the-art in-memory distributed system, which is typically faster than the diskbased distributed one, however, requires a lot of machines for handling billion-scale graphs. In this paper, we propose an I/O efficient parallel algorithm for billion-scale graphs in a single PC. We first propose the Disk-based Sequential access-oriented Parallel processing (DSP) model that exploits sequential disk access in terms of disk I/Os and parallel processing in terms of computation. We then propose an ultra-fast disk-based parallel algorithm for computing connected components, DSP-CC, which largely improves the performance through sequential disk scan and page-level cache-conscious parallel processing. Extensive experimental results show that DSP-CC 1) computes connected components in billion-scale graphs using the limited memory size whereas in-memory algorithms can only support medium-sized graphs with the same memory size, and 2) significantly outperforms all distributed competitors as well as a representative disk-based parallel method. | - |
dc.language | English | - |
dc.publisher | IEEE | - |
dc.title | DSP-CC-: I/O Efficient Parallel Computation of Connected Components in Billion-Scale Networks | - |
dc.type | Conference | - |
dc.identifier.wosid | 000361245300006 | - |
dc.identifier.scopusid | 2-s2.0-84941554873 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 2658 | - |
dc.citation.endingpage | 2671 | - |
dc.citation.publicationname | IEEE International Conference on Data Engineering ICDE | - |
dc.identifier.conferencecountry | FI | - |
dc.identifier.conferencelocation | helsinki | - |
dc.identifier.doi | 10.1109/TKDE.2015.2419665 | - |
dc.contributor.localauthor | Kim, Min-Soo | - |
dc.contributor.nonIdAuthor | Lee, Sangyeon | - |
dc.contributor.nonIdAuthor | Han, Wook-Shin | - |
dc.contributor.nonIdAuthor | Park, Himchan | - |
dc.contributor.nonIdAuthor | Lee, Jeong-Hoon | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.