Run-time spatial resource management for real-time applications on heterogeneous MPSoCs

Abstract

Abstract-Design-time application mapping is limited to a predefined set of applications and a static platform. Resource management at run-time is required to handle future changes in the application set, and to provide some degree of fault tolerance, due to imperfect production processes and wear of materials. This paper concerns resource allocation at run-time, allowing multiple real-time applications to run simultaneously on a heterogeneous MPSoC. Low-complexity algorithms are required, in order to respond fast enough to unpredictable execution requests. We present a decomposition of this problem into four phases. The allocation of tasks to specific locations in the platform is the main contribution of this work. Experiments on a real platform show the feasibility of this approach, with execution times in tens of milliseconds for a single allocation attempt.

Adaptive resource allocation for streaming applications

“…In this approach, the required effort highly depends on the structure and demands of the application graph, where dense graphs result in more computational challenging problems. The time complexity is then in the order of the elements times the number of tasks squared (O(E × T 2 )) [25].…”

“…Our resource allocation heuristic [25] requires an initial step that builds upon the assumption that in each application, a (small) subset of tasks can only be mapped to specific resources in the platform. This assumption holds for tasks responsible for the I/O of the application.…”

On-line dependability enhancement of multiprocessor SoCs by resource management

“…When applications can be started and stopped by externally triggered events, the resource management problem is a non-clairvoyant scheduling problem, for which no optimal solutions can be guaranteed [19]. Therefore, most online resource management approaches use heuristics to solve this complex problem [20], [21].…”

“…Therefore, most online resource management approaches use heuristics to solve this complex problem [20], [21]. In reference [19], the heuristics are explained used to build the resource manager described in this paper.…”

Test standards reuse for structured and cost-efficient dependability management of system-on-chips

“…The SDM computes the updated core voltage and frequency pairs using a control method such as a PID controller. In the case that the required adaptation for a certain core is below the application performance requirements, the SDM considers that this core is not fit to perform those applications, and notifies the application agent to start a dynamic task remapping [Bra10] to another more reliable core.…”