TY - GEN
T1 - Scheduling end-to-end flexible resource reservation requests for multiple end sites
AU - Shi, Li
AU - Sharma, Sushant
AU - Katramatos, Dimitrios
AU - Yu, Dantong
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2015/3/26
Y1 - 2015/3/26
N2 - Wide area research and education networks, such as ESnet and Internet2 in the US and GEANT in Europe, have recently deployed software that makes possible to reserve bandwidth in the form of dynamic circuits. Such circuits offer guaranteed QoS to specific data flows, significantly increasing the reliability and predictability of data transfers. In this paper, we study the problem of constructing routes and scheduling bandwidth reservations for data transfers between multiple pairs of end sites. We develop an algorithm, called RRM, to solve this problem. Our objective is to maximize the number of satisfied data transfer requests while minimizing the total data transfer times. We further prove that our problem is NP-hard and compare our algorithm with a baseline FCFS algorithm through simulations. The simulations indicate that our algorithm accommodates up to 160% more requests and achieves up to 50% shorter average data transfer times than the baseline algorithm.
AB - Wide area research and education networks, such as ESnet and Internet2 in the US and GEANT in Europe, have recently deployed software that makes possible to reserve bandwidth in the form of dynamic circuits. Such circuits offer guaranteed QoS to specific data flows, significantly increasing the reliability and predictability of data transfers. In this paper, we study the problem of constructing routes and scheduling bandwidth reservations for data transfers between multiple pairs of end sites. We develop an algorithm, called RRM, to solve this problem. Our objective is to maximize the number of satisfied data transfer requests while minimizing the total data transfer times. We further prove that our problem is NP-hard and compare our algorithm with a baseline FCFS algorithm through simulations. The simulations indicate that our algorithm accommodates up to 160% more requests and achieves up to 50% shorter average data transfer times than the baseline algorithm.
UR - http://www.scopus.com/inward/record.url?scp=84928023478&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84928023478&partnerID=8YFLogxK
U2 - 10.1109/ICCNC.2015.7069450
DO - 10.1109/ICCNC.2015.7069450
M3 - Conference contribution
AN - SCOPUS:84928023478
T3 - 2015 International Conference on Computing, Networking and Communications, ICNC 2015
SP - 810
EP - 816
BT - 2015 International Conference on Computing, Networking and Communications, ICNC 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2015 International Conference on Computing, Networking and Communications, ICNC 2015
Y2 - 16 February 2015 through 19 February 2015
ER -