Journal article
Queue assignment for fixed-priority real-time flows in time-sensitive networks: Hardness and algorithm
Journal of systems architecture, Vol.116, p.102141
06/01/2021
DOI: 10.1016/j.sysarc.2021.102141
Abstract
Time sensitive networks (TSNs) enable deterministic real-time communication over Ethernet networks. According to IEEE 802.1Qbv standards, TSN switches use gates between queues and their corresponding egress ports to facilitate timing-deterministic communications. Management of switch resources, such as queues, has a significant impact on the schedulability of real-time flows. In this paper, we look into the theoretical foundation of queue management in TSN switches. We prove that the queue assignment problem for realtime flows on time sensitive networks under static priority scheduling is NP-hard in the strong sense, even if the number of queues per port is 3. Then we formulate the problem as a satisfiability modulo theories (SMT) specification. Besides, we propose a worst case response time analysis and a fast heuristic algorithms by eliminating scheduling conflicts. Experiments with randomly generated workload demonstrate the effectiveness of our algorithms for queue assignment of real-time flows.
Details
- Title: Subtitle
- Queue assignment for fixed-priority real-time flows in time-sensitive networks: Hardness and algorithm
- Creators
- Yuhan Lin - Northeastern UniversityXi Jin - Shenyang Institute of AutomationTianyu Zhang - Northeastern UniversityMeiling Han - Nanjing University of Posts and TelecommunicationsNan Guan - Hong Kong Polytechnic UniversityQingxu Deng - Northeastern University
- Resource Type
- Journal article
- Publication Details
- Journal of systems architecture, Vol.116, p.102141
- Publisher
- Elsevier
- DOI
- 10.1016/j.sysarc.2021.102141
- ISSN
- 1383-7621
- eISSN
- 1873-6165
- Number of pages
- 11
- Grant note
- 62072085 / National Natural Science Foundation of China; National Natural Science Foundation of China (NSFC) XLYC1902017 / LinoNing Revitalization Talents Program 2018YFB1702003 / National Key Research and Development Program of China; National Key Research & Development Program of China
- Language
- English
- Date published
- 06/01/2021
- Academic Unit
- Computer Science
- Record Identifier
- 9984696725402771
Metrics
1 Record Views