Hypervisor Resource Management Overview

The hypervisor provides availability guarantees to guest operating systems. For example, multiple server operating systems that have been consolidated onto a single physical computer should not be able to prevent each other from making progress. A user should be able to run a partition that provides telephony or DVR support such that this partition continues to operate irrespective of the potentially adversarial actions of other partitions. The hypervisor addresses only the memory management aspects of this problem.

In accordance with the overall microkernel architecture of the hypervisor, the hypervisor moves support for specific resource policy types, such as quotas or reserves, from the hypervisor into guests. The hypervisor provides minimal support for resource management in connection with several hypercalls that consume memory. In this instance, consume means that the hypercall function might need to allocate memory for internal data structures and that this memory remains allocated beyond the duration of the hypercall.

The resource-management state affects, and is affected by, any hypercall that allocates or frees memory in the hypervisor. Whether and how a particular hypercall function interacts with the resource management state is described in the Remarks section of the reference page for each function. For example, the reference page for the HvCreatePartition function specifies from whose memory pool the initial resources that are required to set up a partition will be deducted.

 

 

Send comments about this topic to Microsoft

Build date: 11/16/2013

Show:
© 2014 Microsoft. All rights reserved.