Jan 24, 2012
PrePrint: Algebraic Optimization for Processing Graph Pattern Queries in the Cloud
Scalable processing of Semantic Web data has become crucial given the rapid increase in available data. MapReduce platforms like Hadoop are now the de-facto standard for large-scale data processing, but have significant limitations for join-intensive workloads typical in Semantic Web processing. Such workloads produce lengthy MapReduce execution workflows with large amount of I/O, sorting and communication costs
Go here to read the rest:
PrePrint: Algebraic Optimization for Processing Graph Pattern Queries in the Cloud