Adoption of a legacy network management protocol for virtualisation

Abstract

Virtualisation is one of the key concepts that allows for abstraction of hardware and software resources on the cloud. Virtualisation has been employed for all aspects of cloud computing including big data processing. There have been arguments based on recent research that indicate that computational efficiency could be more efficient via virtualisation compared to their physical counterparts. A data centre not only represents physical resources but also the collection of virtualised entities that in essence, form virtual networks. Methods to monitor these virtual entities for attributes such as network traffic, performance, sustainability, etc., usually tend to be deployed on ad hoc basis. Understanding the network related attributes of virtualised entities on the cloud will help take critical decisions on management activities such as timing, selection and migration of virtual machines (VMs). In corporate physical data networks, it could have been achieved using four of the network management functional areas, i.e., performance management, configuration management, accounting management and fault management. This chapter discusses, with examples, at how network management principles could be contextualised with virtualisation on the cloud. In particular, the discussion will be centred on the application of Simple Network Management Protocol (SNMP) for gathering behavioural statistics from each virtualised entity.