TY - GEN
T1 - TeaVaR
T2 - 50th Conference of the ACM Special Interest Group on Data Communication, SIGCOMM 2019
AU - Bogle, Jeremy
AU - Bhatia, Nikhil
AU - Ghobadi, Manya
AU - Menache, Ishai
AU - Bjørner, Nikolaj
AU - Valadarsky, Asaf
AU - Schapira, Michael
N1 - Funding Information: We thank Hari Balakrishnan, Jeff Cox, Kimia Ghobadi, Arpit Gupta, Srikanth Kandula, Praveen Kumar, Hongqiang Liu, the anonymous SIGCOMM reviewers, and our shepherd Michael Mitzenmacher. This work was partially supported by NSF grant CNS-1563826 and the Israel Science Foundation. Publisher Copyright: © 2019 Association for Computing Machinery.
PY - 2019/8/19
Y1 - 2019/8/19
N2 - To keep up with the continuous growth in demand, cloud providers spend millions of dollars augmenting the capacity of their widearea backbones and devote significant effort to efficiently utilizing WAN capacity. A key challenge is striking a good balance between network utilization and availability, as these are inherently at odds; a highly utilized network might not be able to withstand unexpected traffic shifts resulting from link/node failures. We advocate a novel approach to this challenge that draws inspiration from financial risk theory: leverage empirical data to generate a probabilistic model of network failures and maximize bandwidth allocation to network users subject to an operator-specified availability target. Our approach enables network operators to strike the utilizationavailability balance that best suits their goals and operational reality. We present TeaVaR (Traffic Engineering Applying Value at Risk), a system that realizes this risk management approach to traffic engineering (TE). We compare TeaVaR to state-of-the-art TE solutions through extensive simulations across many network topologies, failure scenarios, and traffic patterns, including benchmarks extrapolated from Microsoft's WAN. Our results show that with TeaVaR, operators can support up to twice as much throughput as state-ofthe- art TE schemes, at the same level of availability.
AB - To keep up with the continuous growth in demand, cloud providers spend millions of dollars augmenting the capacity of their widearea backbones and devote significant effort to efficiently utilizing WAN capacity. A key challenge is striking a good balance between network utilization and availability, as these are inherently at odds; a highly utilized network might not be able to withstand unexpected traffic shifts resulting from link/node failures. We advocate a novel approach to this challenge that draws inspiration from financial risk theory: leverage empirical data to generate a probabilistic model of network failures and maximize bandwidth allocation to network users subject to an operator-specified availability target. Our approach enables network operators to strike the utilizationavailability balance that best suits their goals and operational reality. We present TeaVaR (Traffic Engineering Applying Value at Risk), a system that realizes this risk management approach to traffic engineering (TE). We compare TeaVaR to state-of-the-art TE solutions through extensive simulations across many network topologies, failure scenarios, and traffic patterns, including benchmarks extrapolated from Microsoft's WAN. Our results show that with TeaVaR, operators can support up to twice as much throughput as state-ofthe- art TE schemes, at the same level of availability.
KW - Availability
KW - Network optimization
KW - Traffic engineering
KW - Utilization
UR - http://www.scopus.com/inward/record.url?scp=85072275367&partnerID=8YFLogxK
U2 - https://doi.org/10.1145/3341302.3342069
DO - https://doi.org/10.1145/3341302.3342069
M3 - Conference contribution
T3 - SIGCOMM 2019 - Proceedings of the 2019 Conference of the ACM Special Interest Group on Data Communication
SP - 29
EP - 43
BT - SIGCOMM 2019 - Proceedings of the 2019 Conference of the ACM Special Interest Group on Data Communication
Y2 - 19 August 2019 through 23 August 2019
ER -