Systems, methods, and media for transparently optimizing a workload of a containment abstraction are provided herein. Methods may include monitoring a workload of the containment abstraction, the containment abstraction being at least partially hardware bound, the workload corresponding to resource utilization of the containment abstraction, converting the containment abstraction from being at least partially hardware bound to being entirely central processing unit (CPU) bound by placing the containment abstraction in a memory store, based upon the workload, and allocating the workload of the containment abstraction across at least a portion of a data center to optimize the workload of the containment abstraction.