Runtime Infrastructure Optimisation in Cloud IaaS Structures

The requirement for elasticity involves the ability of cloud datacenters to add or remove resources at a fine grain and with a lead time of seconds, closely matching resources to the actual demand conditions. Elasticity techniques can assist cloud stakeholders in regards to: (i) alleviating datacenter capital and operating costs, (ii) keeping cloud services continuously available, (iii) supporting market flexibility. To date, most infrastructure scaling methodologies can provide resource reconfiguration decisions to maintain quality properties under environment changes. However, issues related to the timeliness of reconfiguration decisions under dynamic changes are not adequately addressed. In this paper, we describe the PhD motivation, research questions and methodology towards developing cloud infrastructure and hosted services QoS optimisations under environment uncertainties.