Cost-effective social network data placement and replication using graph-partitioning

Hourieh Khalajzadeh, Dong Yuan, John Grundy, Yun Yang

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearch

6 Citations (Scopus)


Social network users are connected based on shared interests, ideas, association with different groups, etc. Common social networks such as Facebook and Twitter have hundreds of millions or even billions of users scattered all around the world sharing interconnected data. Users demand low latency access to not only their own data but also their friends' data. However, social network service providers wish to pay as less as possible to store all data items to meet users' data access latency requirement. Geo-distributed cloud services with virtually unlimited capabilities are suitable for large scale social networks data storage in different geographical locations. Key problems including how to optimally store and replicate these huge data items and how to distribute the requests to different datacentres are addressed in this paper. A novel graph-partitioning based approach is proposed to find a near-optimal data placement of replicas to minimise monetary cost while satisfying the latency requirement. Experiments on a Facebook dataset demonstrate our technique's effectiveness in outperforming other representative placement and replication strategies.

Original languageEnglish
Title of host publicationProceedings - 2017 IEEE 1st International Conference on Cognitive Computing - ICCC 2017
EditorsPaul P. Maglio, Wu Chou
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages8
ISBN (Print)9781538620083
Publication statusPublished - 7 Sep 2017
Externally publishedYes
EventIEEE International Conference on Cognitive Computing 2017 - Honolulu, United States of America
Duration: 25 Jun 201730 Jun 2017
Conference number: 1st (Proceedings)


ConferenceIEEE International Conference on Cognitive Computing 2017
Abbreviated titleICCC 2017
CountryUnited States of America
Internet address


  • Data placement
  • Data replication
  • Graph-partitioning
  • Latency
  • Social network
  • Storage cost

Cite this