In order to make a difference in transaction- oriented distributed environments, performance management efforts must be focused on those components that play a critical role in end-to-end system performance. This paper overviews a flexible approach to identifying these components and quantifying their criticality, through a statistical extension to the classic critical path algorithm for DAGs. A preliminary case study in a real-world service-oriented Grid is used to demonstrate the approach.
Discussion(0)
No comments yet. Be the first to comment.