TY - GEN
T1 - FlowToss
T2 - 20th Annual IFIP Networking Conference, IFIP Networking 2021
AU - Bhatia, Randeep
AU - Lakshman, T. V.
AU - Ozkoc, Mustafa F.
AU - Panwar, Shivendra
N1 - Funding Information:
ACKNOWLEDGMENTS This work was supported in part by NYU Wireless, an Ernst Weber Fellowship, and by the NY State Center for Advanced Technology in Telecommunications (CATT).
Publisher Copyright:
© 2021 IFIP.
PY - 2021/6/21
Y1 - 2021/6/21
N2 - Motivated by important industrial automation use cases, such as closed loop motion control and autonomous mobile robots, we study wait-free scheduling of periodic flows with stringent delay and jitter requirements in time sensitive networks. The goal is to assign initial transmission time-slots to periodic flows so that network queuing delays are eliminated or are very small. We make use of Bézour's Identity to develop simple and fast scheduling algorithms for this NP-hard problem. Operating in an online mode, our algorithms can quickly allocate contention free start time-slots to new flows, without changing allocations of already scheduled flows. Our main results are greedy and random scheduling algorithms that can trade speed for solution quality. Our simulations on different network topologies show that these algorithms are computationally efficient and can easily schedule a large number of flows, thus meeting the requirements of many industrial automation use cases.
AB - Motivated by important industrial automation use cases, such as closed loop motion control and autonomous mobile robots, we study wait-free scheduling of periodic flows with stringent delay and jitter requirements in time sensitive networks. The goal is to assign initial transmission time-slots to periodic flows so that network queuing delays are eliminated or are very small. We make use of Bézour's Identity to develop simple and fast scheduling algorithms for this NP-hard problem. Operating in an online mode, our algorithms can quickly allocate contention free start time-slots to new flows, without changing allocations of already scheduled flows. Our main results are greedy and random scheduling algorithms that can trade speed for solution quality. Our simulations on different network topologies show that these algorithms are computationally efficient and can easily schedule a large number of flows, thus meeting the requirements of many industrial automation use cases.
KW - Industrial Automation
KW - Scheduling Algorithms
KW - Time Sensitive Networking
KW - Zero Queuing
UR - http://www.scopus.com/inward/record.url?scp=85112800390&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85112800390&partnerID=8YFLogxK
U2 - 10.23919/IFIPNetworking52078.2021.9472838
DO - 10.23919/IFIPNetworking52078.2021.9472838
M3 - Conference contribution
AN - SCOPUS:85112800390
T3 - 2021 IFIP Networking Conference, IFIP Networking 2021
BT - 2021 IFIP Networking Conference, IFIP Networking 2021
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 21 June 2021 through 24 June 2021
ER -