Computational Science - ICCS 2004

A 5-year-profiling in production mode at the University of Stuttgart has shown that more than 40% of the execution time of Message Passing Interface (MPI) routines is spent in the collective communication routines MPI Allreduce and MPI Reduce. Although MPI implementations are now available for about 10 years and all vendors are committed to this Message Passing Interface standard, the vendors’ and publicly available reduction algorithms could be accelerated with new algorithms by a factor between 3 (IBM, sum) and 100 (Cray T3E, maxloc) for long vectors. This paper presents five algorithms optimized for different choices of vector size and number of processes. The focus is on bandwidth dominated protocols for power-of-two and non-power-of-two number of processes, optimizing the load balance in communication and computation.

[1]  Bamshad Mobasher,et al.  Using Ontologies to Discover Domain-Level Web Usage Profiles , 2002 .

[2]  Jaideep Srivastava,et al.  Automatic personalization based on Web usage mining , 2000, CACM.

[3]  Kyungsook Han,et al.  A fast layout algorithm for protein interaction networks , 2003, Bioinform..

[4]  Andrew McCallum,et al.  Building Domain-Specific Search Engines with Machine Learning Techniques , 1999 .

[5]  O. Ronneberger,et al.  Automated pollen recognition using gray scale invariants on 3D volume image data , 2000 .

[6]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[7]  Ping Li,et al.  Pollen texture identification using neural networks , 1999 .

[8]  GeunSik Jo,et al.  Collaborative Information Filtering by Using Categorized Bookmarks on the Web , 2001, INAP.

[9]  Jaideep Srivastava,et al.  Data Preparation for Mining World Wide Web Browsing Patterns , 1999, Knowledge and Information Systems.

[10]  R. Albert,et al.  The large-scale organization of metabolic networks , 2000, Nature.

[11]  Myra Spiliopoulou,et al.  The Impact of Site Structure and User Environment on Session Reconstruction in Web Usage Analysis , 2002, WEBKDD.

[12]  Timothy W. Finin,et al.  Yahoo! as an ontology: using Yahoo! categories to describe documents , 1999, CIKM '99.

[13]  Wei-Ying Ma,et al.  A unified framework for Web link analysis , 2002, Proceedings of the Third International Conference on Web Information Systems Engineering, 2002. WISE 2002..

[14]  G. E. Taylor,et al.  Computerized identification of pollen grains by texture analysis , 1990 .

[15]  Jaideep Srivastava,et al.  Web mining: information and pattern discovery on the World Wide Web , 1997, Proceedings Ninth IEEE International Conference on Tools with Artificial Intelligence.