efficient multi-way theta-join processing using mapreduce研究生英文论文.pdfVIP

Efficient Multiway ThetaJoin Processing Using MapReduce Xiaofei Zhang Lei Chen Min Wang HKUST HKUST HP Labs China Hong Kong Hong Kong Beijing, China zhangxf@cse.ust.hk leichen@cse.ust.hk min.wang6@ ABSTRACT challenging problem along with the development of database Multi-way Theta-join queries are powerful in describing com- technology. Early works, like [8][26][22] and etc., have elab- plex relations and therefore widely employed in real prac- orated the complexity of the problem and presented their tices. However, existing solutions from traditional distribut- evaluation strategies. However, their solutions do not scale ed and parallel databases for multi-way Theta-join queries to process the multi-way Theta-joins over the data of tremen- cannot be easily extended to fit a shared-nothing distributed dous volumes. For instance, as reported from Facebook [5] computing paradigm, which is proven to be able to sup- and Google [11], the underlying data volume is of hundreds port OLAP applications over immense data volumes. In of tera-bytes or even peta-bytes. In such scenarios, solu- this work, we study the problem of efficient processing of tions from the traditional distributed or parallel databases multi-way Theta-join queries using MapReduce from a cost- are infeasible due to unsatisfactor