Provided by: University of Antwerp
Date Added: Jul 2011
Cloud computing can be defined as an abstraction of services from infrastructures (i.e. hardware), platforms and applications (i.e. software) by virtualization of resources. This paper describes two approaches to scaling in cloud environments - semi-automatic (also called by request) and automatic (also called on demand) - and explains why the latter is to prefer. Semi-automatic scaling is illustrated with an example of Amazon Elastic Compute Cloud, whereas automatic scaling is illustrated with an example of Amazon Web Services Elastic Beanstalk.