TY - GEN
T1 - CLOVE
T2 - 15th ACM Workshop on Hot Topics in Networks, HotNets 2016
AU - Katta, Naga
AU - Hira, Mukesh
AU - Ghag, Aditi
AU - Kim, Changhoon
AU - Keslassy, Isaac
AU - Rexford, Jennifer L.
N1 - Publisher Copyright:
© 2016 ACM.
PY - 2016/11/9
Y1 - 2016/11/9
N2 - Multi-tenant datacenters predominantly use equal-cost multipath (ECMP) routing to distribute traffic over multiple network paths. However, ECMP static hashing causes unequal load-balancing and collisions, leading to low throughput and high latencies. Recently proposed alternatives for load-balancing perform better, but are impractical as they require either changing the tenant VM network stacks (e.g., MPTCP) or replacing all the network switches (e.g., CONGA). In this paper, we argue that the end-host hypervisor provides a sweet spot for implementing a spectrum of load-balancing algorithms that are fine-grained, congestionaware, and reactive to network dynamics at round-trip timescales. We propose CLOVE, a scalable hypervisorbased load-balancer that requires no changes to guest VMs or to physical network switches. CLOVE uses standard ECMP in the physical network, learns about equal-cost network paths using a traceroute mechanism, and learns about congestion state along these paths using standard switch features such as ECN. It then manipulates packet header fields in the hypervisor virtual switch to route traffic over less congested paths. We introduce different variants of CLOVE that differ in the way they learn about congestion in the physical network. Using extensive simulations, we show that CLOVE captures some 80% of the performance gain of best-of-breed hardware-based load-balancing algorithms without the need for expensive hardware replacement.
AB - Multi-tenant datacenters predominantly use equal-cost multipath (ECMP) routing to distribute traffic over multiple network paths. However, ECMP static hashing causes unequal load-balancing and collisions, leading to low throughput and high latencies. Recently proposed alternatives for load-balancing perform better, but are impractical as they require either changing the tenant VM network stacks (e.g., MPTCP) or replacing all the network switches (e.g., CONGA). In this paper, we argue that the end-host hypervisor provides a sweet spot for implementing a spectrum of load-balancing algorithms that are fine-grained, congestionaware, and reactive to network dynamics at round-trip timescales. We propose CLOVE, a scalable hypervisorbased load-balancer that requires no changes to guest VMs or to physical network switches. CLOVE uses standard ECMP in the physical network, learns about equal-cost network paths using a traceroute mechanism, and learns about congestion state along these paths using standard switch features such as ECN. It then manipulates packet header fields in the hypervisor virtual switch to route traffic over less congested paths. We introduce different variants of CLOVE that differ in the way they learn about congestion in the physical network. Using extensive simulations, we show that CLOVE captures some 80% of the performance gain of best-of-breed hardware-based load-balancing algorithms without the need for expensive hardware replacement.
UR - http://www.scopus.com/inward/record.url?scp=85002213857&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85002213857&partnerID=8YFLogxK
U2 - 10.1145/3005745.3005751
DO - 10.1145/3005745.3005751
M3 - Conference contribution
AN - SCOPUS:85002213857
T3 - HotNets 2016 - Proceedings of the 15th ACM Workshop on Hot Topics in Networks
SP - 155
EP - 161
BT - HotNets 2016 - Proceedings of the 15th ACM Workshop on Hot Topics in Networks
PB - Association for Computing Machinery, Inc
Y2 - 9 November 2016 through 10 November 2016
ER -