No abstract

Matrix languages, including MATLAB and Octave, are established standards for applications in science and engineering. They provide interactive programming environments that are easy to use due to their scripting languages with matrix data types. Current implementations of matrix languages do not fully utilise high-performance, special-purpose chip architectures such as the IBM PowerXCell processor (Cell), which is currently used in the fastest computer in the world.We present a new framework that extends Octave to harness the computational power of the Cell. With this framework the programmer is relieved of the burden of introducing explicit notions of parallelism. Instead the programmer uses a new matrix data-type to execute matrix operations in parallel on the synergistic processing elements (SPEs) of the Cell. We employ lazy evaluation semantics for our new matrix data-type to obtain execution traces of matrix operations. Traces are converted to data dependence graphs; operations in the data dependence graph are lowered (split into sub-matrices), scheduled and executed on the SPEs. Thereby we exploit (1) data parallelism, (2) instruction level parallelism, (3) pipeline parallelism and (4) task parallelism of matrix language programs. We conducted extensive experiments to show the validity of our approach. Our Cellbased implementation achieves speedups of up to a factor of 12 over code run on recent Intel Core2 Quad processors.

In this paper we introduce a novel system that comprises techniques advancing the query processing in wireless sensor networks. Our system facilitates time triggered queries that are scheduled in a distributed fashion among sensor nodes. Thus, time synchronisation is of paramount importance. Since accurate time synchronisation requires more energy, our system allows a trade off between precision of time and energy according to the user requirements. To minimize the communication overhead for query processing, our system employs new query execution mechanisms. We have implemented our query processing system on Sun TM Small Programmable Object Technology (SPOT) sensor network platform. The system was entirely programmed in Java enabling an object oriented design and implementation. It provides a friendly graphical user interface for query management and visualisation of results.

Web search engine: characteristics of user behaviors and their implication Science in China Series F-Information Sciences 44, 351 (2001); A Web-based visual analytics system for real estate data SCIENCE CHINA Information Sciences 56, 052112 (2013); PREDAV-H1: a user-friendly web server for predicting antigenic variants of influenza H1N1 viruses SCIENCE CHINA Life Sciences 62, 426 (2019); Detecting micro-blog user interest communities through the integration of explicit user relationship and implicit topic relations SCIENCE CHINA Information Sciences 60, 092105 (2017);. RESEARCH PAPER .

No abstract