Temporal workload-aware replicated partitioning for social networks

Simple item page
buir.contributor.authorAykanat, Cevdet
dc.citation.epage14en_US
dc.citation.issueNumber11en_US
dc.citation.spage1en_US
dc.citation.volumeNumber26en_US
dc.contributor.authorTurk, A.en_US
dc.contributor.authorSelvitopi, R. O.en_US
dc.contributor.authorFerhatosmanoglu, H.en_US
dc.contributor.authorAykanat, Cevdeten_US
dc.date.accessioned2015-07-28T12:02:37Z
dc.date.available2015-07-28T12:02:37Z
dc.date.issued2014-11en_US
dc.departmentDepartment of Computer Engineeringen_US
dc.description.abstractMost frequent and expensive queries in social networks involve multi-user operations such as requesting the latest tweets or news-feeds of friends. The performance of such queries are heavily dependent on the data partitioning and replication methodologies adopted by the underlying systems. Existing solutions for data distribution in these systems involve hash- or graph-based approaches that ignore the multi-way relations among data. In this work, we propose a novel data partitioning and selective replication method that utilizes the temporal information in prior workloads to predict future query patterns. Our method utilizes the social network structure and the temporality of the interactions among its users to construct a hypergraph that correctly models multi-user operations. It then performs simultaneous partitioning and replication of this hypergraph to reduce the query span while respecting load balance and I/O load constraints under replication. To test our model, we enhance the Cassandra NoSQL system to support selective replication and we implement a social network application (a Twitter clone) utilizing our enhanced Cassandra. We conduct experiments on a cloud computing environment (Amazon EC2) to test the developed systems. Comparison of the proposed method with hash- and enhanced graph-based schemes indicate that it significantly improves latency and throughput.en_US
dc.description.provenanceMade available in DSpace on 2015-07-28T12:02:37Z (GMT). No. of bitstreams: 1 8274.pdf: 1033988 bytes, checksum: da5339cb43a4d59199ec7ec57554f636 (MD5)en
dc.identifier.doi10.1109/TKDE.2014.2302291en_US
dc.identifier.issn1041-4347
dc.identifier.urihttp://hdl.handle.net/11693/12689
dc.language.isoEnglishen_US
dc.publisherInstitute of Electrical and Electronics Engineersen_US
dc.relation.isversionofhttp://dx.doi.org/10.1109/TKDE.2014.2302291en_US
dc.source.titleIEEE Transactions on Knowledge & Data Engineeringen_US
dc.subjectCassandraen_US
dc.subjectSocial network partitioningen_US
dc.subjectSelective replicationen_US
dc.subjectReplicated hypergraph partitioningen_US
dc.subjectTwitteren_US
dc.subjectNosqlen_US
dc.titleTemporal workload-aware replicated partitioning for social networksen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
8274.pdf
Size:
1009.75 KB
Format:
Adobe Portable Document Format
Download

Collections

Scholarly Publications - Computer Engineering