Fair Online Power Capping for Emergency Handling in Multi-Tenant Cloud Data Centers

Abstract

In view of the high capital expense for scaling up power capacity to meet the escalating demand, maximizing the utilization of built capacity has become a top priority for multi-tenant data center operators, where many cloud providers house their physical servers. The traditional power provisioning guarantees a high availability, but is very costly and results in a significant capacity under-utilization. On the other hand, power oversubscription improves utilization but offers no availability guarantees due to the necessity of power reduction to handle the resulting power emergencies. Given these limitations, we propose a novel hybrid power provisioning approach, called HyPP, providing a combination of two different power availabilities to tenants: capacity with a very high availability, plus additional capacity with a medium availability that may be unavailable for up to a certain amount during each billing period. For HyPP, we design an online algorithm to coordinate tenants power reduction at runtime when the tenants aggregate power demand exceeds capacities. Our algorithm aims at achieving long-term fairness in tenants power reduction. We analyze the theoretical performance and derive a good competitive ratio in terms of fairness compared to the offline optimum. We also validate our algorithm through simulations under realistic settings.