VirtualRack: Bandwidth-aware virtual network allocation for multi-tenant datacenters

Huang, Tianlin; Chen, Rong; Tang, Yazhe; Hu, Chengchen; Li, Jinming; Zhang, Peng

doi:10.1109/icc.2014.6883883

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Oktopus [Ballani et al, 2011] abstracts VDCs into virtual clusters, which are VMs connected to central virtual switch in a star topology. VirtualRack [Huang et al, 2014] and HVC-ACE [Rost et al, 2015] use a lessrestricted hose-model [Duffield et al, 1999] abstraction for VDCs. NETSOLVER is the first sound and complete VDC allocation approach that supports arbitrary VDC and data center topologies.…”

Section: Related Workmentioning

confidence: 99%

“…The key insight to our approach is that VDC allocation can be formulated in terms of monotonic theories, a recently developed concept that lies at the heart of the SMT solver MONOSAT [Bayless et al, 2015]. Exploiting this insight, we constructed NETSOLVER, a virtual data center allocation procedure that is scalable, sound, and complete, with support Algorithm Complete Multi-path VDC Topology Data Center Topology SecondNet [Guo et al, 2010] All All Importance Sampling [Tantawi, 2012] All Tree Oktopus [Ballani et al, 2011] Star All VDCPlanner [Zhani et al, 2013] All All HVC-ACE [Rost et al, 2015] Hose All VirtualRack [Huang et al, 2014] Hose All Z3-AR [Yuan et al, 2013] All Tree NETSOLVER (this work) All All Table 1: Characteristics of sound VDC allocation algorithms from the literature and the NETSOLVER approach introduced in this work.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Scalable Constraint-based Virtual Data Center Allocation

Bayless

Kodirov

Beschastnikh

et al. 2017

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence

View full text Add to dashboard Cite

Constraint-based techniques can solve challenging problems arising from highly diverse applications. This paper considers the problem of virtual data center (VDC) allocation, an important, emerging challenge for modern data center operators. To solve this problem, we introduce NETSOLVER, which is based on the general-purpose constraint solver MONO-SAT. NETSOLVER represents a major improvement over existing approaches: it is sound, complete, and scalable, providing support for end-to-end, multipath bandwidth guarantees across all the layers of hosting infrastructure, from servers to top-of-rack switches to aggregation switches to access routers. NETSOLVER scales to realistic data center sizes and VDC topologies, typically requiring just seconds to allocate VDCs of 5-15 virtual machines to physical data centers with 1000+ servers, maintaining this efficiency even when the data center is nearly saturated. In many cases, NETSOLVER can allocate 150%−300% as many total VDCs to the same physical data center as previous methods. Essential to our solution efficiency is our formulation of VDC allocation using monotonic theories, illustrating the practical value of the recently proposed SAT modulo monotonic theories approach.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%