Cyclo-Static Data Flow Model for TDM

摘要

Time Division Multiplexing (TDM) allows resource sharing amongst the tasks of concurrent applications, where each application may have its own end-to-end hard real time requirements. Current data flow modeling techniques for TDM arbitrated tasks with cyclo-static execution times are over-pessimistic in modeling their worst-case temporal behavior. This causes unnecessary over-reservation of resources to the application, leading to under-utilization of system resources and unnecessary rejection of additional applications. More accurate models for representing TDM exist but have restrictive assumptions on execution times (should be static) and/or the allocation of resources. We propose a conservative data flow model that accurately estimates the worst-case temporal behavior of TDM arbitrated tasks with cyclo-static execution as well as for the general case. This enables optimized resource allocation for TDM arbitration. Quantitatively, we show that our new model provides up to 30% improvement of resource allocation, in a case study of a wireless LAN radio.