Decreasing Utilization of Systems with Multi-Rate Cause-Effect Chains While Reducing End-to-End Latencies
The Logical Execution Time (LET) model has deterministic properties which dramatically reduce the complexity of analyzing temporal requirements of multi-rate cause-effect chains. The configuration (length and position) of task's communication intervals directly define which task instances propagate data through the chain and affect end-to-end latencies. Since not all task instances propagate data through the chain, the execution of these instances wastes processing resources. By manipulating the configuration of communication intervals, it is possible to control which task instances are relevant for data propagation and end-to-end latencies. However, since tasks can belong to more than one cause-effect chain, the problem of configuring communication intervals becomes non-trivial given the large number of possible configurations. In this paper, we present a method to decrease the waste of processing resources while reducing end-to-end latencies. We use a search algorithm to analyze different communication interval configurations and find the combination that best decrease system utilization while reducing end-to-end latencies. By controlling data propagation by means of precedence constraints, our method modifies communication intervals and controls which task instances affect end-to-end latencies. Despite the sporadic release time of some task instances during the analysis, our method transforms those instances into periodic tasks. We evaluate our work using synthetic task sets and the automotive benchmark proposed by BOSCH for the WATERS industrial challenge.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Lyapunov Characterization for ISS of Impulsive Switched Systems
In this study, we investigate the ISS of impulsive switched systems that have modes with both stable and unstable flows. We assume that the switching signal satisfies mode-dependent average dwell and leave time condition…
PipeMare: Asynchronous Pipeline Parallel DNN Training
Pipeline parallelism (PP) when training neural networks enables larger models to be partitioned spatially, leading to both lower network communication and overall higher hardware utilization. Unfortunately, to preserve t…
RECUR: Resource Exhaustion Attack via Recursive-Entropy Guided Counterfactual Utilization and Reflection
Large Reasoning Models (LRMs) employ reasoning to address complex tasks. Such explicit reasoning requires extended context lengths, resulting in substantially higher resource consumption. Prior work has shown that advers…
Kraken: Inherently Parallel Transformers For Efficient Multi-Device Inference
Large Transformer networks are increasingly used in settings where low inference latency can improve the end-user experience and enable new applications. However, autoregressive inference is resource intensive and requir…
GPULanguage ModelingLanguage ModellingPredictive Vehicle Repositioning for On-Demand Ride-Pooling Services
On-Demand Ride-Pooling services have the potential to increase traffic efficiency compared to private vehicle trips by decreasing parking space needed and increasing vehicle occupancy due to higher vehicle utilization an…