Parallel Distributed and Cloud Computing for Databases

In the PDCCDB project, we are designing efficient algorithms for the evaluation of relational queries in share-nothing parallel and distributed system. The algorithms are analysed using structured models of parallelism and distribution. An important technique in our work is the use of histograms that allow to design efficient strategies for communications and to avoid skews.

Team Members

• Mostafa Bamha (LIFO, University of Orléans)
• Mohamad Al Hajj Hassan (EXQIM / LIFO, University of Orléans)
• Frédéric Loulergue (LIFO, University of Orléans)

Past members: F. Bentayeb, M. Exbrayat, G. Hains

Publications

2011

• Mohamad Al Hajj Hassan, Mostafa Bamha, and Frédéric Loulergue. An Efficient Skew-insensitive Algorithm for Join Processing on Grid Architectures. In 5th ACM SIGPLAN workshop on High-Level Parallel Programming and Applications, 2011, to appear

2010

• Mohamad Al Hajj Hassan & Mostafa Bamha, ​Semi-join Computation on Distributed File Systems Using Map-Reduce-Merge Model, in (SAC'2010), Proceedings of the 25th ACM Symposium On Applied Computing. Sierre, Suisse. Pp. 406-413, ACM Press. 2010

2009

• Mohamad Al Hajj Hassan & Mostafa Bamha, An Efficient Pipelined Parallel Join Algorithm on Heterogeneous Distributed Architectures, in Software and Data Technologies, Extended and revised ICSFOT'2008 Best papers Book.. Pp. 119-133, Springer Berlin Heidelberg. 2009
• Mohamad Al Hajj Hassan & Mostafa Bamha, ​An Efficient Parallel Algorithm for Evaluating Join Queries on Heterogeneous Distributed Systems, in (HiPC'09), 16th Annual International Conference on High Performance Computing, pp. 350-358, IEEE Computer Society. 2009

2008

• Mohamad Al Hajj Hassan & Mostafa Bamha, Parallel Processing of Group-By Join Queries on Shared Nothing Machines, in Software and Data Technologies, Extended and revised -ICSOFT'2006 Best papers- Book. pp. 230-241, Springer Berlin Heidelberg. 2008
• Mohamad Al Hajj Hassan & Mostafa Bamha, Pipelined Parallelism in Multi-Join Queries on Heterogeneous Shared Nothing Architectures, in (ICSOFT'2008), Proceedings of the Third International Conference on Software and Data Technologies. Porto, Portugal, Vol. Volume PL/DPS/KEICSOFT (PL/DPS/KE). Pp. 127-134, INSTICC Press. 2008

2007

• Mohamad Al Hajj Hassan & Mostafa Bamha, An Optimal Evaluation of Group-By-Join Queries in Distributed Architectures, Proceedings of the International Conference on Web Information Systems and Technologies. Barcelona, Espagne. 2007

2006

• Mohamad Al Hajj Hassan & Mostafa Bamha, Parallel Processing Of Group-By Join Queries On Shared Nothing Machines, in (ICSOFT'06), Proceedings of the International Conference on Software Data Technologies. Setubal, Portugal. pp. 301-307, INSTICC press. 2006

2005

• M. Bamha, ​An optimal and skew-insensitive join and multi-join algorithm for ditributed architectures, in Proceedings of the International Conference on Database and Expert Systems Applications (DEXA'2005). 22-26 August, Copenhagen, Danemark, Vol.3588, Lecture Notes in Computer Science, pp. 616-625, Springer-Verlag, 2005
• M. Bamha et G. Hains, ​An Efficient Equi-semi-join Algorithm for Distributed Architectures, in the International Conference on Computational Science (ICCS'2005). 22-25 May, Atlanta, USA, Vol.3515, Lecture Notes in Computer Science, pp.755-763, Springer-Verlag, 2005

2003

• M. Bamha et M. Exbrayat, Pipelined parallelism for multi-join queries on shared nothing machines, in Proceedings of the international conference on Parallel Computing (PARCO 2003), Dresden, Germany, 02-05 September, 2003
• M. Bamha et M. Exbrayat, ​Pipelining a Skew-Insensitive Parallel Join Algorithm, Parallel Processing Letters" journal, Vol.13, No.3, 2003

2001

• M. Bamha et G. Hains, Frequency-adaptive join for Shared Nothing machines, Parallel and Distributed Computing Practices, Nova Science Publishers, Vol.2, No.3, pp.333-345, 2001

2000

• M. Bamha et G. Hains, ​A skew-insensitive algorithm for join and multi-join operations on Shared Nothing machines, in 11th International Conference on Database and Expert Systems Applications, DEXA'2000, M. Ibrahim, J. Küng et N. Revell ed., No.1873, Lecture Notes in Computer Science, pp.644-653, Greenwich, U.K., Springer-Verlag, September, 2000

1999

• M. Bamha, F. Bentayeb et G. Hains, ​An efficient scalable parallel view maintenance algorithm for shared nothing multi-processor machines, in 10th International Conference on Database and Expert Systems Applications, DEXA'99, Trevor Bench-Capon, Giovanni Soda et A Min Tjoa ed., No.1677, Lecture Notes in Computer Science, pp.616-625, Florence, Italy, Springer-Verlag, 1999

Funding

Work on cloud computing and databases is partly funded in the ​INEX project.