Data protection should be a key concern for every one of your customers today. If just one of their virtual machines (VMs) were to fail for any reason, chances are that it would have a detrimental impact on their business – in the most extreme cases it can even put the company out of business.
“Hybrid cloud might be all the rage today, but the multi-cloud model — where IT deploys a mix of cloud services from different providers — is next in line.”
SearchCloudComputing, Multi-cloud strategies to follow hybrid cloud craze
As we journey from private to public cloud adoption, stopping at hybrid in the middle, we are now looking at the next evolution – a multi-cloud approach. Not all clouds are the same and the need for multiple clouds is a reality. So the natural next step for DevOps is to adopt a multi-cloud strategy.
The majority of enterprises are looking into the deployment of multiple clouds for both the short and long-term according to survey results reported in Cloud Tech News. The research found that:
Say you have deployed a multi-tier web application in the cloud, with each tier made up of a few servers. How large does each tier need to be? That depends on the demand placed on the application at any given time, and in many cases this is hard, if not impossible, to predict. Each tier may have its own internal bottlenecks and will have to scale differently from the others. But in any case, it all boils down to:
Too little server resource and your application user’s experience will suffer;
Too much and you are over provisioning and over spending on infrastructure.