In recent times, we have access to huge volumes of data. The reason for this explosion of data can be attributed to social media. Since the data availability is exploding, the complexity of processing is also increasing. Every organisation will have its own private/internal cloud for data processing. All the requests from the user and the internal requests will be processed in this private cloud. The private cloud will be configured according to the organisational needs and it will have a limit for processing the requests. If the private cloud runs out of resources, the organisation can burst the additional workload to the public/external cloud on an on-demand basis. This process is known as cloud bursting.
Cloud bursting becomes extremely crucial in catering to the varying workload levels. An internal cloud will never have a capacity larger than the average workload of the organisation. But there is no guarantee that the workload will forever be within the average limit. Spikes are quite common. Few days it may be more and few other days it may be even more. To overcome this, the concept of cloud bursting is used. You may wonder why the capacity of internal cloud is never increased! The reason for this is the additional cost involved in doing so and these spikes come and go. So any organisation will try to keep the resources limited and control costs.-
Cloud bursting is not as easy as it is described here. As mentioned already, it is available on on-demand basis. Private vendors will manage external data centres and make it available for organizations on demand. There are many challenges involved in this process. Since many organisations will be using the same external cloud, there may be security concerns and it is possible for the external cloud to run out of resources. The main challenges in this regard are:
Cloud bursting will show the way for organizations to optimize storage costs while taking care of data security. Companies will make the best use of infrastructures by launching resources on demand and managing workloads in the best way.