Summary: The study is examining the field data from private cloud systems over a three years period. The main focus of the study is to analyze approximately 22 PB data from 8000 physical data boxes and draw conclusions over the data volume, velocity and storage demands. The study also examines the dependency over virtual machines and other resource allocation. The study takes in to consideration a variety of software application including web, database, files and mails.
Strengths: The study uses physical storage devices in to consideration that provides it a dependency on actual data instead of virtually created scenarios. The study surveys 90,000 VMs in this regard. The study is a longitudinal in nature that is spread over three years to yield productive results. Statistical techniques have been effectively used to draw meaningful inferences.
Weaknesses: The data set used in the study does not consist of information about cache size. It has used only memory size and the calculations are based on the correlation between memory size and cache size. There is an 11% duplication among boxes the reason behind which is dubious in the research study.
Questions: What are the differences between the results of the present study and other studies carried out earlier?
The study has been carried out on private cloud datacenters, what are the disadvantages of private cloud data centers for such researches?