As multiprocessor software solutions find their way into critical applications such as avionics as a means of increasing performance and reliability, the need for both debugging and testing tools increases dramatically. The authors focus on the gathering of accurate run-time information providing support for both debugging and testing of embedded multiprocessor systems by using run-time identification of simple events and postexecution identification of compound events. Additionally, they propose a way of combining specification-based (black-box) system testing and automatic monitoring to substantially reduce the complexity, tedium, and cost of complete structural (white-box) testing.>

An O(n/sup 2/) heuristic algorithm using randomized decisions is developed for determining a near-optimal set of nodes. For small values of n, the authors can determine how close the heuristic solution is to the optimal set of nodes. They also compare their heuristic to other algorithms reported in the literature.>

This paper summarizes the execution cost of join operations performed by parallel executing processors. Four different parallel join algorithms are proposed for execution on multiple processing nodes interconnected on a tree structured communication network. Secondary storage is accessed in parallel by leaf nodes. An average execution cost analysis is presented for the multiple processor join methods. For a reasonable ratio of result to operand cardinality, joins which use hashing for semi-join of operands at the leaf node secondary storage interface are predicted to perform better than nested-loop and sort-merge joins. Both node memory capacity and join result cardinality have a large influence on the relative performance of the join methods. This analysis method forms a basis for selecting among alternative processing methods for statically linked, multiple processor computer architectures.

Comparison of three multiprocessor computer architectures for database support is made possible through evaluation of response time expressions. These expressions are derived by parameterizing algorithms performed by each machine to execute a relational algebra query. Parameters represent properties of the database and components of the machines. Studies of particular parameter values exhibit response times for conventional machine technology, for low selectivity, high duplicate occurrence, and parallel disk access, increasing the number of processors, and improving communication and processing technology.