Accurate run-time performance prediction for multi-application multiprocessor systems
Akash S. Kumar, Bart Mesman, Henk Corporaal, Yajun Ha · 2008
Abstract — Non-preemptive multi-processor platforms are increasingly being developed to support the performance requirements of modern systems with multiple applications. Due to a huge number of possible combinations of these multiple applications, it becomes a challenge to predict their performance in advance. This becomes even more important when applications may be dynamically started and stopped in the system. Misprediction may result in reduced quality of applications and lower the user-experience. Since modern embedded systems allow users to download and add applications at run-time, a complete designtime analysis is not possible. In this paper, we present a technique to accurately predict the performance of applications at run-time before they execute in the system. The technique uses performance expressions computed off-line from the application specifications. A run-time iterative