Efficient Weighted Histogramming on GPUs with HASH

Top-k retrieval is at the core of many modern applications: from large scale web search and advertising platforms, to text extenders and content management systems. In these systems, queries are evaluated using two major families of algorithms: document-at-a-time (DAAT) and term-at-a-time (TAAT). DAAT and TAAT algorithms have been studied extensively in the research literature. In this talk, I’ll present an analysis and comparison of several DAAT and TAAT algorithms, focusing on the performance characteristics of these algorithms. Copyright is held by the author/owner(s). LSDS-IR’13, February 5, Rome, Italy. BIOGRAPHY Marcus Fontoura has finished his Ph.D. studies in 1999, at the Pontifical Catholic University of Rio de Janeiro, Brazil (PUC-Rio) in a joint program with the Computer Systems Group, University of Waterloo, Canada. Since then he held research posts at the Princeton University Computer Science Department, IBM Almaden Research Center, and Yahoo! Research. Currently he is a Research Scientist and Member of Technical Staff at Google. His main areas of research in the last years have been Web Search, Computational Advertising, Enterprise search, and Databases. He has more than 40 published papers and 20 issued patents. His complete CV is available at: http://fontoura.org. INVITED SPEAKER: Quasi-succinct Indices Sebastiano Vigna Univ. degli Studi di Milano Italy [email protected]