In the last years, prominent advances in software virtualization (e.g., Hypervisor) have enabled the agile deployment of cloud services and, at the same time, server consolidation operations through agile Virtual Machine (VM) operations (e.g., start-up, snapshot, cloning, and migrations) across Data Center (DC) servers. On the other hand, such advances pose new challenges in the overall management of the DC infrastructure since VM operations may occur much more frequently in virtualized than in legacy DCs and produce not only a computational load on the involved servers, but also a network load on all the switches connecting the servers. In this context, the Software-Defined Network (SDN) paradigm opens up new opportunities to design convergent resource management systems which is able to address the provisioning of cloud services while meeting adequate network performance during VM provisioning and migrations. In this paper, a promising advanced monitoring and resource allocation function for SDN-based DC networks has been proposed aiming at reporting possible suitable paths connecting specified servers that can be exploited for VM migrations. In such a way, network-assisted migrations can be triggered that assure adequate VM migration delays according to specified requirements (e.g., minimum delay for live VM migration) while addressing different management targets (e.g., network load balancing, minimization of energy consumption).

Orchestration algorithms for network-assisted virtual machine migrations using OpenDayLight Controller

Martini B;
2015-01-01

Abstract

In the last years, prominent advances in software virtualization (e.g., Hypervisor) have enabled the agile deployment of cloud services and, at the same time, server consolidation operations through agile Virtual Machine (VM) operations (e.g., start-up, snapshot, cloning, and migrations) across Data Center (DC) servers. On the other hand, such advances pose new challenges in the overall management of the DC infrastructure since VM operations may occur much more frequently in virtualized than in legacy DCs and produce not only a computational load on the involved servers, but also a network load on all the switches connecting the servers. In this context, the Software-Defined Network (SDN) paradigm opens up new opportunities to design convergent resource management systems which is able to address the provisioning of cloud services while meeting adequate network performance during VM provisioning and migrations. In this paper, a promising advanced monitoring and resource allocation function for SDN-based DC networks has been proposed aiming at reporting possible suitable paths connecting specified servers that can be exploited for VM migrations. In such a way, network-assisted migrations can be triggered that assure adequate VM migration delays according to specified requirements (e.g., minimum delay for live VM migration) while addressing different management targets (e.g., network load balancing, minimization of energy consumption).
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12606/2618
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
social impact