An Evaluation of Alternative Physical Graph Data Designs for Processing Interactive Social Networking Actions
Abstract
This study quantifies the tradeoff associated with alternative physical representations of a social graph for processing interactive social networking actions. We conduct this evaluation using a graph data store named Neo4j deployed in a client-server (REST) architecture using the BG benchmark. In addition to the average response time of a design, we quantify its SoAR defined as the highest observed throughput given the following service level agreement: 95 % of actions to observe a response time of 100 ms or faster. For an action such as computing the shortest distance between two members, we observe a tradeoff between speed and accuracy of the computed result. With this action, a relational data design provides a significantly faster response time than a graph design. The graph designs provide a higher SoAR than a relational one when the social graph includes large member profile images stored in the data store.
References
- 1.Amsden, Z., Bronson, N., Cabrera III, G., Chakka, P., Dimov, P., Ding, H., Ferris, J., Giardullo, A., Hoon, J., Kulkarni, S., Lawrence, N., Marchukov, M., Petrov, D., Puzar, L., Venkataramani, V.: TAO: how facebook serves the social graph. In: SIGMOD Conference (2012)Google Scholar
- 2.Angles, R., Boncz, P., Larriba-Pey, J., Fundulaki, I., Neumann, T., Erling, O., Neubauer, P., Martinez-Bazan, N., Kostev, V., Toma, I.: The Linked data benchmark council: a graph and RDF industrybenchmarking effort. SIGMOD Rec. 43, 27–31 (2014)CrossRefGoogle Scholar
- 3.Angles, R., Prat-Pérez, A., Dominguez-Sal, D., Larriba-Pey, J.: Benchmarking database systems for social network applications. In: First International Workshop on Graph Data Management Experiences and Systems, GRADES 2013 (2013)Google Scholar
- 4.Armstrong, T., Ponnekanti, V., Borthakur, D., Callaghan, M.: LinkBench: a database benchmark based on the facebook social graph. In: ACM SIGMOD, June 2013Google Scholar
- 5.Bai, X., Junqueira, F.P, Silberstein, A.: Cache refreshing for online social news feeds. In: CIKM (2013)Google Scholar
- 6.Barahmand, S.: Benchmarking interactive social networking actions. Ph.D. thesis, Computer Science Department, USC (2014)Google Scholar
- 7.Barahmand, S., Ghandeharizadeh, S.: BG: a benchmark to evaluate interactive social networking actions. In: Proceedings of 2013 CIDR, January 2013Google Scholar
- 8.Barahmand, S., Ghandeharizadeh, S.: Benchmarking correctness of operations in big data applications. In: Proceedings of IEEE MASCOTS (2014)Google Scholar
- 9.Barahmand, S., Ghandeharizadeh, S.: Extensions of BG for testing and benchmarking alternative implementations of feed following. In: ACM SIGMOD Workshop on Reliable Data Services and Systems (RDSS), June 2014Google Scholar
- 10.Barahmand, S., Ghandeharizadeh, S., Yap, J.: A comparison of two physical data designs for interactive social networking actions. In: CIKM (2013)Google Scholar
- 11.Boncz, P.: LDBC: benchmark for graph and RDF data management. In: IDEAS, October 2013Google Scholar
- 12.Transaction Processing Performance Council. TPC Benchmarks. http://www.tpc.org/information/benchmarks.asp
- 13.Nishtala, R., et al.: Scaling memcache at Facebook. In: NSDI (2013)Google Scholar
- 14.Ghandeharizadeh, S., Barahmand, S.: A mid-flight synopsis of the BG social networking benchmark. In: Rabl, T., Raghunath, N., Poess, M., Bhandarkar, M., Jacobsen, H.-A., Baru, C. (eds.) WBDB 2013. LNCS, vol. 8585, pp. 19–31. Springer, Heidelberg (2013)CrossRefGoogle Scholar
- 15.Gray, J.: The Benchmark Handbook for Database and Transaction Systems, 2nd edn. Morgan Kaufmann, San Mateo (1993). ISBN 1055860-292-5zbMATHGoogle Scholar
- 16.Holzschuher, F., Peinl, R.: Performance of graph query languages: comparison of cypher, gremlin and native access in Neo4J. In: Proceedings of the Joint EDBT/ICDT 2013 Workshops, EDBT 2013 (2013)Google Scholar
- 17.Junqueira, F.P., Leroy, V., Serafini, M., Silberstein, A.: Shepherding social feed generation with sheep. In: SNS (2012)Google Scholar
- 18.Labouseur, A., Olsen, P., Hwang, J: Scalable and robust management of dynamic graph data. In: VLDB Workshop on Big Dynamic Distributed Data (2013)Google Scholar
- 19.Sears, R., Ingen, C.V., Gray, J.: To BLOB or not to BLOB: large object storage in a database or a filesystem. Technical report MSR-TR-2006-45, Microsoft Research (2006)Google Scholar
- 20.Silberstein, A., Machanavajjhala, A., Ramakrishnan, R.: Feed following: the big data challenge in social applications. In: DBSocial (2011)Google Scholar
- 21.Silberstein, A., Terrace, J., Cooper, B.F., Ramakrishnan, R.: Feeding frenzy: selectively materializing users’ event feeds. In: SIGMOD Conference (2010)Google Scholar
- 22.The Neo4j Team. The Neo4j Manual V2.1.1, 29 May 2014. http://www.neo4j.org
- 23.Tesoriero, C.: Getting Started with OrientDB. Packt Publishing Ltd, Birmingham (2013)Google Scholar