A New Framework for Join Product Skew

Different types of data skewness can result in load imbalance in the context of parallel joins under the shared nothing architecture. We study one important type of skewness, join product skew (JPS). A static approach based on frequency classes is proposed which takes for granted the data distribution of join attribute values. It comes from the observation that since the join selectivity can be expressed as a sum of products of frequencies of the join attribute values, an appropriate assignment of join sub-tasks, that takes into consideration the magnitude of the frequency products can alleviate the join product skew. Motivated by the aforementioned ascertainment, we propose an algorithm, called Handling Join Product Skew (HJPS), to handle join product skew. Key words: Parallel DBMS, join operation, join selectivity, data distribution, data skew, load imbalance, shared nothing architecture

• Type: Journal
• Year: 2010
• Where: CORR
• Authors: Foto N. Afrati, Victor Kyritsis, Paraskevas V. Lekeas, Dora Souliou
• Keywords: CORR 2010, Education, Join Product, Load Imbalance, Product Skew