Server scale-out is an approach to take advantage of mass produced enterprise 
computing building blocks to build high-end systems. Servers scale-out 
provisioning, on the other hand, refers to the ability to allocate, schedule, 
scale and manage compute and I/O resources of multiple platforms to efficiently
 and positively react to runtime changes including workload changes and faults.
 The main objective of such provisioning is to reduce the Total Cost of 
Ownership (TCO) through improving resource utilization, server availability, 
security, manageability and reducing the overhead associated with 
serviceability & maintainability.

In this talk, I'll present couple approaches to provisioning. The first is 
based on launching/terminating/resizing virtual machines running on separate 
physical servers to react to runtime changes. The second is based on 
partitioning and aggregating compute and I/O resources.