Abstract

In this paper, we present the design and implementation of POSSE, a new, lightweight computational steering system based on a client/server programming model. We demonstrate the effectiveness of this software system by illustrating its use for a visualization client designed for a particularly demanding real-time application--wake-vortex simulations for multiple aircraft running on a parallel Beowulf cluster. We describe how POSSE is implemented as an object-oriented, class-based software library and illustrate its ease of use from the perspective of both the server and client codes. We discuss how POSSE handles the issue of data coherency of distributed data structures, data transfer between different hardware representations, and a number of other implementation issues. Finally, we consider how this approach could be used to augment AVOSS (an air traffic control system currently being developed by the FAA) to significantly increase airport utilization while reducing the risks of accidents. Document type: Part of book or chapter of book

Abstract

Traditional Worst-Case Execution-Time (WCET) analysis is very complex. It has to deal with path analysis, to identify and describe the possible execution paths through the code to be analyzed, and it has to model the worst-case timing of the possible paths on the target hardware. The latter is again nontrivial due to interference of modern hardware features like instruction pipelines, caches, and parallel instruction-execution units on the processor. Document type: Part of book or chapter of book