In this paper parallel algorithms for fractal image coding on MIMD architectures are introduced and discussed. It turns out that the crucial point for the choice of a suitable parallelization strategy is the memory capacity of a processor element. Experimental results show a linear speedup of the proposed algorithms.

web hosting service, web server cluster, load balancing, scalability, super linear speedup, performance analysis FLEX is a new cost effective, "locality aware" load balancing solution for a shared web hosting service implemented on a cluster of machines [C99]. FLEX allocates hosted web sites to different machines in the cluster based on the sites' processing and memory requirements which are es...

Our randomized preprocessing of a matrix by means of augmentation counters its degeneracy and ill conditioning, uses neither pivoting nor orthogonalization, readily preserves matrix structure and sparseness, and leads to dramatic speedup of the solution of general and structured linear systems of equations in terms of both estimated arithmetic time and observed CPU time.

Workstations connected by local area networks are now common. Using these resources to speed up processor intensive tasks is an important area of study. This paper demonstrates an almost linear speedup as workstations are added to the task of determining the N-th prime integer.

This report details the implementation of a fragment extraction algorithm using an average case linear time tree kernel. Given a treebank, the algorithm extracts all fragments that occur at least twice, along with their frequency. Evaluation shows a -fold speedup over a quadratic fragment extraction implementation. Additionally, we add support for trees with discontinuous constituents.

The parallelization of the Computational Fluid Dynamics chimera based flow solver XAIR through the utilization of Parallel Virtual Machine software is described. The required modifications are discussed in a four step process. Efficiency results indicate an almost linear speedup and limited communication overhead for a near perfect test case.