Links as a Service (LaaS): Guaranteed Tenant Isolation in the Shared Cloud

Abstract

The most demanding tenants of shared clouds require complete isolation from their neighbors, in order to guarantee that their application performance is not affected by other tenants. Unfortunately, while shared clouds can offer an option whereby tenants obtain dedicated servers, they do not offer any network provisioning service, which would shield these tenants from network interference.In this paper, we introduce Links as a Service (LaaS), a new abstraction for cloud service that provides isolation of network links. Each tenant gets an exclusive set of links forming a virtual fat-tree, and is guaranteed to receive the exact same bandwidth and delay as if it were alone in the shared cloud. Consequently, each tenant can use the forwarding method that best fits its application. Under simple assumptions, we derive theoretical conditions for enabling LaaS without capacity over-provisioning in fat-trees. New tenants are only admitted in the network when they can be allocated hosts and links that maintain these conditions. LaaS is implementable with common network gear, tested to scale to large networks and provides full tenant isolation at the worst cost of a 10% reduction in the cloud utilization.

Bouncer: A Resource-Aware Admission Control Scheme for Cloud Services

“…Cloud computing has the potential to transform a large part of the IT industry and to make the software even more attractive as a service as well as to shape the way IT hardware is designed, utilized, and purchased [1]. The success of any cloud management software critically depends on the flexibility, scale, and efficiency with which it can utilize the underlying hardware resources while providing necessary performance isolation [2]. Therefore, resource management at the cloud-scale requires the management platform to provide a rich set of resource controls that can balance the quality of service (QoS) of tenants with the overall resource efficiencies of the datacenters.…”

Mitigating Inter-Job Interference via Process-Level Quality-of-Service

“…However, adaptive routing does not necessarily reduce contention significantly [1]. Researchers have also studied job placement strategies that minimize contention [3], [23], [24], but these approaches typically reduce machine utilization.…”

Efficient routing and reconfiguration in virtualized HPC environments with vSwitch‐enabled lossless networks

“…[2][3][4]25 Others have focused in cloud tenant isolation in the link level, 26,27 whereas Ranadive, in his doctoral dissertation, dedicated his efforts on virtualized resource management with lossless networks. [2][3][4]25 Others have focused in cloud tenant isolation in the link level, 26,27 whereas Ranadive, in his doctoral dissertation, dedicated his efforts on virtualized resource management with lossless networks.…”

“…In addition, in the presence of multiple tenant groups in the network, techniques like tenant-aware routing can be integrated with the proposed routing scheme to provide network-wide isolation among tenants 26,27. The value of the weight reflects the proportion of the vSwitch to leaf switch link capacity allocated to a VM.…”

“…The scheme can be combined with per VM rate limits enforcement on each vSwitch to ensure that a VM is not allowed to exceed its allocated capacity. In addition, in the presence of multiple tenant groups in the network, techniques like tenant-aware routing can be integrated with the proposed routing scheme to provide network-wide isolation among tenants 26,27.…”