hp home products & services support solutions how to buy
hp logo - invent
corner hp labs corner
search search
contact hp contact hp
hp labs home hp labs home
about hp labs about hp labs
research research
news and events news and events
careers @ labs careers @ labs
technical reports technical reports
talks and speeches talks and speeches
worldwide sites worldwide sites
corner corner
Capacity Management and Demand Prediction for Next Generation Data Centers

Many enterprises are beginning to exploit shared resource pool environments. In these environments, the application workloads exploit a common set of hardware resources. These are complex environments where selecting per-application scheduler parameter settings is a challenging task. It is a challenge because i) the capacity of resource pools are generally overbooked, i.e., the sum of per-application peak demands are greater than the capacity of the pool, and ii) because different applications can have different quality of service (QoS) requirements that are affected by the applications' ability to obtain capacity when needed.

We are designing and developing a capacity management service for resource pools that support enterprise applications. It automatically searches for workload assignments that best meet objectives such as consolidation, load leveling, and problem solving. Consolidation packs workloads onto a small set of resources, load leveling distributes workloads across a fixed size resource pool to lessen the likelihood of service level violations, and problem solving recommends alternative assignments to lessen the impact of unexpected demands on a particular resource. The service evaluates alternative workload assignments more quickly and accurately than human operators. As a result it can lower administrative costs, reduce the risks associated with resource sharing, and improve the use of resource pool infrastructure.

Related Papers and Reports

printing icon
printing instructions printing instructions
Privacy Statement Legal Notices © 1994-2001 Hewlett-Packard Company