For lighting simulations in architecture there is the need for correct illumination calculation of virtual scenes. The Radiance Synthetic Imaging System delivers an excellent solution to that problem. Unfortunately, simulating complex scenes leads to long computation times even for one frame. This paper proposes a parallelization strategy which is suited for scenes with medium to high complexit...

The Graphics Environment for Max, or GEM, is a real-time computer graphics language which operates under the SGI version of Max. GEM supports abstract computer graphics as well as image and pixel manipulations, enabling the user to control the graphics as objects, instead of as a sequence of vertices. Because it maintains the Max paradigm, GEM is an easy to use development environment.

In this paper we present a parallel algorithm for solving the inverse eigenvalue problem for real symmetric Toeplitz matrices. The algorithm is implemented using the parallel library ScaLAPACK, based on the communication model MPI. Experimental results are reported on a SGI Power Challenge and on a cluster of PC's connected by a Myrinet network. The results show that the algorithm is portable a...

This paper explains how the Cray MTA-2 might be a good platform on which to run an existing suite of finite element codes. We discuss the issues involved and compare the performance on the MTA with that on an SGI Origin 3800.

Complete Swiss glacier inventories are available for 1850 (reconstructed) and 1973 (from aerial photography). Connected to the GLIMS project a new Swiss glacier inventory for approximately the year 2000 (SGI 2000) is compiled mainly based on satellite imagery. The developed and applied remote sensing and GIS methods are described in part I of the contribution. In part II, the inventory design, ...

Scalability and reliability are inseparable in high-performance computing. Fault-isolation through hardware is a popular means of providing reliability. Unfortunately, such isolation also increases communication latencies: typically, one has to drop into and out of the kernel to communicate between failure domains. On the other hand, relaxing fault isolation domains allows e cient communication...

Matrix multiplication is widely used in a variety of applications and is often one of the core components of many scientific computations which includes graph theory, numerical methods, digital control and signal processing. Multiplication of large matrices require a lot of computation time as its complexity is O(n), where n is the dimension of the matrix. A serial algorithm to compute large ma...

Recursive algorithms are generally not straightforward to be processed in parallel, particularly when an octree is involved. Some researchers have tried to process an octree in parallel, but this solution is restricted to recursive subdivision methods that guarantee surface containment in an octant. In this paper we show a whole class of recursive subdivision algorithms which does not fall into...