Profit Optimization in SLA-Aware Cloud Services with a Finite Capacity Queuing Model

Table 1

Cost and system parameters.

Parameter	Description
𝐶𝑅	Server provisioning cost per server per unit time
𝜀	Power consumption cost per Watt per unit time
𝜇𝑏	A baseline service rate
𝜇	A given execution rate (𝜇 > 𝜇𝑏)
𝜇^′	An increased execution rate (𝜇 − 𝜇𝑏)
𝐶𝐻	Cost incurred by holding tasks in buffer per unit time
𝐶𝑊	Cost incurred by tasks waiting in buffer per unit time
𝐶𝑈	Cost incurred by providing per waiting buffer space per unit time