One of the first issues that companies encounter as they begin migrating their Oracle workloads to the cloud is that they find they need to overprovision the number of cloud resources – including compute, memory, network bandwidth, capacity, and performance – that they need to invest in. This is because all these resources are directly tied together, meaning you can’t get more of one without getting more of the others.
One resource that Oracle workloads will desperately need in the cloud is ultra-fast performance. To get it, without Silk you’ll most likely need to deploy a much greater number of resources than you actually need – resulting in a significantly higher cloud bill from your provider. We will go into detail about how this works shortly below.