DISSERTATION

Efficient Virtual Machine Mobility in Cloud Computing

Abstract

Cloud computing change the Internet into a new computing platform, is a business model that achieve purchase on-demand and pay-per-use in network, has a broad development prospects. Virtualization technology is a core technology of cloud computing, the virtual machine is the basic unit of the cloud computing platforms, cloud providers providing services to clients by virtual machines must ensure the security and isolation. The capability to migrate applications with no perceivable effect to the end user enables service providers to develop new and improved methods for provisioning and maintaining data center infrastructure. Cloud computing providers can now perform hardware maintenance, consolidate CPU and memory resources, or migrate mission-critical applications from a data center when necessary without affecting the service-level agreements (SLAs) of the applications. A cumbersome issue is dealing with the network deployment context required to preserve end-to-end connectivity as VMs are migrated to new physical machines that are likely located on different networks. To the best of our knowledge, all proposed ideas rely on 'virtual networks' (i.e., overlay networks) to facilitate the interconnection of VMs. We take a radically different approach. The significant growth of Internet traffic and increase of routing tables require solutions to address Internet scalability and resiliency. The approach chosen is to move away from the flat legacy Internet routing to hierarchical routing, separating edge networks from transit networks. In the context of Cloud computing and Internet datacenter, Id/locator splitting provides a convenient method for establishing the network context for a VM that is migrated to a new physical machine. A protocol, the Location Identifier Separation Protocol (LISP) follows this paradigm, adding in particular a control-plane based on a mapping system, which separates the location function from the identification function of the IP address. OpenLISP is an open source implementation of the LISP proposal in the FreeBSD OS. OpenLISP provides complete Data Plane support for both encapsulation and decapsulation operations, for IPv4 and IPv6, as well as EID-to-RLOC mapping storage and efficient lookup. As part of the thesis research, control plane implementation of OpenLISP is done. Assuming transit/edge approach is widely deployed in the future Internet; our approach to live migration eliminates the need for non-standard and complex virtual networking schemes to interconnect the VM platforms. Our solution guarantees optimal routing between clients and the VM end-point that moved, regardless of its location. In addition, the solution does not require any change in the DNS infrastructure, which overall reduces operating expenses for the data center administrator.

Keywords:
Cloud computing Computer science Virtual machine Operating system

Metrics

0
Cited By
0.00
FWCI (Field Weighted Citation Impact)
5
Refs
0.06
Citation Normalized Percentile
Is in top 1%
Is in top 10%

Topics

Cloud Computing and Resource Management
Physical Sciences →  Computer Science →  Information Systems
IoT and Edge/Fog Computing
Physical Sciences →  Computer Science →  Computer Networks and Communications
Distributed and Parallel Computing Systems
Physical Sciences →  Computer Science →  Computer Networks and Communications

Related Documents

JOURNAL ARTICLE

Cost Efficient Virtual Machine Brokering in Cloud Computing

Dong-Ki KangSeong-Hwan KimChan-Hyun Youn

Journal:   KIPS Transactions on Computer and Communication Systems Year: 2014 Vol: 3 (7)Pages: 219-230
JOURNAL ARTICLE

EFFICIENT VIRTUAL MACHINE LOAD BALANCING USING CLOUD COMPUTING ENVIRONMENT

Kavana.KKavya SMK . RajithaRavali.MS Supreeth

Journal:   Zenodo (CERN European Organization for Nuclear Research) Year: 2020
JOURNAL ARTICLE

EFFICIENT ALLOCATION OF VIRTUAL MACHINE IN CLOUD COMPUTING ENVIRONMENT

Sonam Rathore

Journal:   International Journal of Computer Science and Informatics Year: 2013 Pages: 147-150
© 2026 ScienceGate Book Chapters — All rights reserved.