To clarify, are you asking as to how we'll match the performance of our on premise infrastructure to our cloud infrastructure?

If so, we're actually able to realize greater performance from our on premise infrastructure, as its entirely dedicated to running our workloads (vs. shared tenancy in the cloud, where in the underlying bare metal hardware is running multiple virtual machines, for multiple customers).
In our datacenter, we can scale up (add more resources/instance) and out (add more instances) at our discretion, vs being limited by the cloud provider's tiers (fixed resource allocation/tier: ie: memory, CPU, etc...).

Please let us know if this fully addresses your question, thanks!