Download now Free registration required
Shared data centers that use virtualized servers as building blocks (e.g., clouds) are emerging as an exciting technology that provides a computational platform for customers without incurring a steep setup cost. The providers of a shared data center, however, need intelligent mechanisms to deal with bursts in resource requirements as well as failures that may significantly reduce resource availability. In this work, the author investigates mechanisms to ensure that a shared data center can operate within a power budget, while maximizing the overall revenue earned by the provider. The paper presents the BrownMap methodology that is able to ensure that data centers can deal both with outages that reduces the available power or with surges in workload.
- Format: PDF
- Size: 234.1 KB