Hierarchical, competitive scheduling of multiple DAGs in a dynamic heterogeneous environment

With the advent of large-scale heterogeneous environments, there is a need for matching and scheduling algorithms which can allow multiple, directed acyclic graph structured applications to share the computational resources of the network. This paper presents a hierarchical matching and scheduling framework where multiple applications compete for the computational resources on the network. In this environment, each application makes its own scheduling decisions. Thus, no centralized scheduling resource is required. Applications do not need direct knowledge of the other applications - knowledge of other applications arrives indirectly through load estimates (like queue lengths). This paper presents an algorithm, called the dynamic hierarchical scheduling algorithm, which schedules tasks within this framework. A series of simulations are presented to examine the performance of these algorithms in this environment, compared with a more conventional, single-user environment.