Multi-Resource Allocation Control for Fair QoS Management in Real-Time Systems
Fumiko Harada, Toshimitsu Ushio, Yukikazu Nakamoto
Abstract
Fumiko Harada, Toshimitsu Ushio, Yukikazu Nakamoto
Abstract
In this paper, we propose an adaptive resource control method for multiple resource real-time systems. Execution results of applications are evaluated as a QoS level and a fair QoS management is an important issue. The fair QoS level depends on the number of active applications and their characteristics so that it may change dynamically due to the current states of the real-time systems. The proposed controller activates at discrete times and updates resource allocations to achieve the optimal fair QoS level. Moreover, the algorithms used in the controller solve the update in O(mn) time per its activation. We derive sufficient conditions for achievement of the fair resource allocation. Simulation experiments show that a fair resource allocation can be achieved under the conditions.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, we propose an adaptive resource control method for multiple resource real-time systems. Execution results of applications are evaluated as a QoS level and a fair QoS management is an important issue. The fair QoS level depends on the number of active applications and their characteristics so that it may change dynamically due to the current states of the real-time systems. The proposed controller activates at discrete times and updates resource allocations to achieve the optimal fair QoS level. Moreover, the algorithms used in the controller solve the update in O(mn) time per its activation. We derive sufficient conditions for achievement of the fair resource allocation. Simulation experiments show that a fair resource allocation can be achieved under the conditions.
Key concepts: Quality of service, Resource allocation, Computer science, Resource management (computing), Controller (irrigation), Resource (disambiguation), Distributed computing, Control (management)