Back in the history of the cloud (say, eight years ago), there were copious debates around the topic of cloudbursting. Cloudbursting, for those unaware of the term, describes an approach toward hybrid infrastructure whereby a workload could run on-premises for standard load periods and then, when spikes in traffic occurred, would magically “burst” into the cloud for extra capacity.
While many appreciated the idea of cloudbursting from a conceptual viewpoint, others doubted its practicality.
But despite cloudbursting not really coming to pass, the idea of hybrid infrastructure, whereby organizations have workloads across many setups (from on-premises to the public cloud and on to multiple private clouds), has become very much a reality. One of the reasons for this is for risk reduction — the thinking goes that by using a variety of different service providers, critical issues with one provider are, in theory, less likely to have a negative impact on the organization.