Cost of the Network
A hybrid architecture is more than just a simple combination of enterprise resources and cloud resources, and thus its total cost is more than the sum of those two resource types. For example, there may be additional load balancing needed, a cloud operating system or hybrid cloud management software, performance and usage monitoring, each of which adds to the cost of the hybrid and shifts the breakeven point. Perhaps the most important additional cost in implementing a hybrid is that of the network.
Returning to the car example, a hybrid mix of owned cars and rental cars sounds good, but to implement it, we have to have some way of accessing the rental cars. If we happen to live next door to a rental car location, that would be particularly handy, but such a fortuitous situation is unlikely. The optimal cost of implementing a hybrid solution depends on the network costs, which may vary. Moreover, as author and cloud expert Barrie Sosinksy has said, “If any cloud-computing system resource is difficult to plan for, it is network capacity.”9
The total or marginal costs associated with a network are highly dependent on existing network services and infrastructure, the application architecture, and the pattern of demand.
In the best case, the application doesn’t need any additional networking between the dedicated, flat-rate resources and the on-demand, pay-per-use resources. This scenario can arise in three ways:
Get Cloudonomics: The Business Value of Cloud Computing, + Website now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.