Embedded multimedia systems often run multiple time-constrained applications simultaneously. To meet the throughput constraints given in the specification, each application must be provided with enough resources by the underlying architecture, which is generally a multiprocessor system-on-chip (MPSoC). For this purpose, a mechanism for task binding and scheduling is required to provide each application with a timing guarantee, keeping in mind the available resources like processor(s) and memory bandwidth. Commonly, synchronous dataflow graphs (SDFGs) are used to model time-constrained multimedia applications. There are resource allocation strategies for SDFGs that help in formulating efficient techniques for calculating the throughput of a bounded and scheduled SDFG. The strategies are effective in terms of run-time and allocated resources. However, there is no unified modeling technique to simultaneously represent the application and the underlying architecture with resource allocation. This paper discusses a novel modeling technique using Colored Timed Petri Nets (CTPNs), which can be used to model the application as well as the architecture and the resource allocation. Such a representation helps in checking properties like liveness and boundedness for the application, taking into account the resource allocation and thus helping in defining satisfactory schedules for the executable tasks.
Optimal Resource Allocation for Multimedia Applications Offloading in Mobile Edge Computing
