Brems is the Idaho State University Physics Department's Beowulf REsource for Monte Carlo Simulations. Beowulfs present unique engineering challenges because of the many choices in vendors, architectures, and hardware afforded by tapping into commodity server and specialty high performance computing markets. Brems is no exception to this and represents a unique balance of code performance, hard...

We present a new result concerning the parallelisation of DBSCAN, a Data Mining algorithm for density-based spatial clustering. The overall structure of DBSCAN has been mapped to a skeletonstructured program that performs parallel exploration of each cluster. The approach is useful to improve performance on high-dimensional data, and is general w.r.t. the spatial index structure used. We report...

In this paper, we have implemented and evaluated the performance of local predecimation with range index communication parallelization strategy for fractal image compression on a beowulf cluster of workstations. The strategy effectively balances the load among workstations. We have evaluated the execution time of LPRI, varying the number of workstations and user-specified root mean square error...

Storage has been extensively studied during the past few decades (Foster et al., 1997; Jos Guimares, 2001). However, the emerging trends on distributed computing bring new solutions for existent problems. Grid computing proposes a distributed approach for data storing. In this paper, we introduce a Grid-based system (ARCO) developed for multimedia storage of large ammounts of data. The system i...

Fast real-time monitoring of system information is important to the understanding of parallel system especially for a large cluster system that appeared recently. Making the system fast and scalable at the same time is still a challenging task. This paper presents the design and implementation of a fast and real time monitoring system called SCMS/RMS. This system is a part of more comprehensive...

A new algorithm for Monte Carlo calculation of the double exchange model is studied. The algorithm is commonly applicable to wide classes of strongly correlated electron systems which involve itinerant electrons coupled with thermodynamically fluctuating fields. Using moment expansions of the density of states with Chebyshev polynomials, the algorithm provides an efficient calculation on large ...

We examine the scalability and performance of a legacy liquid crystal code on a PC (Beowulf) cluster consisting of 16 dualprocessor Pentium III/450s. This code was originally designed for use on a Unix workstation cluster of less than 8 machines. In particular, we examine the effectiveness of using potentially more efficient techniques such as non-blocking communication calls and whether the us...