Live migration (LM) is dealt with in this paper as a one-time task or function, performed by moving a virtual machine from one physical machine to another, located in the same or a different data center, without interrupting its operation. The paper also introduces readers to disaster recovery (DR) as a set of practices and activities regarding the continuity of operation of the physical and virtual information technology assets of an organization. With more and more companies moving to cloud services, and companies deploying redundant data centers so that businesses can operate 24/7 without failures, the paper is relevant for defining the expectations and needs of data center solutions. The paper defines terms such as RPO (recovery point objective) and RTO (recovery time objective), which are benchmarks for measuring the effectiveness of LM and DR in systems. Because the paper is intended to be a very detailed survey of the landscape comparing companies such as VMWare, Cisco, Netapp, and so on, concepts that optimize the time such as deduplication and compression are surveyed too. There are several research papers that are compared in this paper. Networking terminologies such as BGP multihoming are also surveyed extensively. The paper also surveys many combinations of industry-leading solutions to see how they offer DR options. One example is how Silver Peak solutions can be combined with Netapp’s Snapmirror to provide good DR solutions.
The paper is targeted at three types of professionals: industry executives and product managers who want to see the landscape and make improvements on their existing products; CIOs and IT professionals interested in what solution will be better for their companies’ LM and DR problems; and academics who want to research the existing solutions and come up with completely new paradigms or solutions that will make significant improvements. The paper achieves these objectives by explaining the topic in a very rigid and detailed manner.