Cloud computing and in particular cloud based applications like Google’s Gmail and Dropbox has recently grown in size and user base. The increase in user traffic and
the unpredictable loads, highlights the importance of dynamically allocating cloud based resources when it is needed most. Cloud based resources are described in three layers, namely Infrastructure as a Servive (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS), with each layer building on the next.
My research will aim to apply load prediction and resource provisioning, also referred to as ‘auto-scaling’, on the Platform layer of a cloud. At this level the cloud application can be divided into three modules; the container, the network and the database module. The hypothesis is: Applying load prediction and resource provisioning on each of the platform modules will show an improvement over applying auto-scaling on the cloud infrastructure.
I am a person with a distributed train-of-thought, have tried to memorised Pi’s decimals, likes to re-use code and improve on it and enjoys playing squash and ultimate Frisbee and good food.My contact details: firstname.lastname@example.org