Virtualized management of remote presentation sessions using virtual machines having load above or below thresholds

1. A method for managing a plurality of hosts, the plurality of hosts configured to execute a plurality of virtual machines (VMs), a VM being configured to serve a remote presentation session, comprising:

receiving a request to connect a client to a remote presentation session;

making a first determination that no VM of the plurality of VMs is both set to accept remote presentation session connections and has a load less than a first threshold indicative of a level of computer resource usage above which remote presentation sessions should not be added;

identifying, in response to the first determination, a subset of the plurality of VMs, each VM of the subset being set to a drain state indicative of allowing the VM of the subset to enter a sleep state;

identifying a first VM in the subset having a highest load of VMs in the subset;

setting the first VM to accept remote presentation session connections;

sending the first VM an indication to conduct the remote presentation session with the client;

receiving a request to connect a second remote presentation session; and

sending a second VM an indication to conduct the second remote presentation session upon determining that no VM of the plurality of VMs has a load below the first threshold and determining that a second VM of the plurality of VMs has a lowest load above the first threshold.

2. The method of claim 1 , further comprising:

receiving a request to connect to a second remote presentation session;

determining that a host of the plurality of hosts has a highest load below a second threshold below the first threshold, the host configured to execute a second VM of the plurality of VMs; and

sending the second VM an indication to conduct the second remote presentation session.

3. The method of claim 1 , further comprising:

receiving a request to connect a second remote presentation session;

determining that no VM of the subset of VMs of the plurality of VMs that are set to refuse connections may be set to accept connections;

determining that a second VM of the plurality of VMs can accept a load of the second remote presentation session, the second VM being in a sleeping state;

waking the second VM; and

sending the second VM an indication to conduct the second remote presentation session.

4. The method of claim 1 , further comprising:

receiving a request to connect a second remote presentation session;

determining that the request to connect a second remote presentation session corresponds to a disconnected remote presentation session conducted by a second VM of the plurality of VMs;

determining that the second VM is running; and

sending the second VM an indication to conduct the second remote presentation session.

5. The method of claim 1 , further comprising:

receiving a request to connect a second remote presentation session;

determining that the request to connect to a second remote presentation session corresponds to a disconnected remote presentation session conducted by a second VM of the plurality of VMs;

determining that the second VM is in a saved state;

determining that a physical host of the second VM has available load to execute the second VM;

waking the second VM; and

sending the second VM an indication to conduct the second remote presentation session.

6. The method of claim 1 , further comprising:

receiving a request to connect a second remote presentation session;

determining that the request to connect to a second remote presentation session corresponds to a disconnected remote presentation session conducted by a second VM of the plurality of VMs;

determining that the second VM is in a saved state;

determining that a physical host of the second VM lacks available load to execute the second VM;

identifying a third VM of the plurality of VMs, the identifying based at least in part on the third VM operating on a host having a highest load among hosts in the plurality of hosts that have a load below a second threshold, the second threshold below the first threshold and indicative of running the third VM in a low power state; and

sending the third VM an indication to conduct the second remote presentation session.

7. The method of claim 1 , further comprising:

determining that a state of a second VM of the plurality of VMs is incorrect based on a load of the second VM; and

changing the state of the second VM.

8. The method of claim 7 , further comprising:

receiving an indication of the load of the second VM from a virtual machine monitor (VMM) that manages the execution of the second VM.

9. The method of claim 8 , wherein the VMM monitors the load of the second VM based on a system call made from the second VM to a host on which the second VM and the VMM execute.

10. The method of claim 8 , wherein the VMM monitors the load of the second VM by querying the second VM for the load.

11. A system for managing a plurality of hosts, the plurality of hosts configured to execute a plurality of virtual machines (VMs), a VM being configured to serve a remote presentation session, comprising:

a processor; and

a memory bearing processor executable instructions that, upon execution by the processor, cause the system at least to:

receive a request to connect a client to a remote presentation session;

make a first determination that no VM of the plurality of VMs is both set to accept remote presentation session connections and has a load less than a first threshold indicative of a level of computer resource usage above which remote presentation sessions should not be added;

identify, in response to the first determination, a subset of the plurality of VMs, each VM of the subset being set to a drain state indicative of allowing the VM of the subset to enter a sleep state;

identify a first VM in the subset having a highest load of VMs in the subset;

set the first VM to accept remote presentation session connections;

send the first VM an indication to conduct the remote presentation session with the client;

receive a request to connect a second remote presentation session; and

send a second VM an indication to conduct the second remote presentation session upon determining that no VM of the plurality of VMs has a load below the first threshold and determining that a second VM of the plurality of VMs has a lowest load above the first threshold.

12. The system of claim 11 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

receive a request to connect to a second remote presentation session;

determine that a host of the plurality of hosts has a highest load below a second threshold below the first threshold, the host configured to execute a second VM of the plurality of VMs; and

send the second VM an indication to conduct the second remote presentation session.

13. The system of claim 11 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

receive a request to connect a second remote presentation session;

determine that no VM of the subset of VMs of the plurality of VMs that are set to refuse connections may be set to accept connections;

determine that a second VM of the plurality of VMs can accept a load of the second remote presentation session, the second VM being in a sleeping state;

wake the second VM; and

send the second VM an indication to conduct the second remote presentation session.

14. The system of claim 11 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

receive a request to connect a second remote presentation session;

determine that the request to connect a second remote presentation session corresponds to a disconnected remote presentation session conducted by a second VM of the plurality of VMs;

determine that the second VM is running; and

send the second VM an indication to conduct the second remote presentation session.

15. The system of claim 11 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

receive a request to connect a second remote presentation session;

determine that the request to connect to a second remote presentation session corresponds to a disconnected remote presentation session conducted by a second VM of the plurality of VMs;

determine that the second VM is in a saved state;

determine that a physical host of the second VM has available load to execute the second VM;

wake the second VM; and

send the second VM an indication to conduct the second remote presentation session.

16. The system of claim 11 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

receive a request to connect a second remote presentation session;

determine that the request to connect to a second remote presentation session corresponds to a disconnected remote presentation session conducted by a second VM of the plurality of VMs;

determine that the second VM is in a saved state;

determine that a physical host of the second VM lacks available load to execute the second VM;

identify a third VM of the plurality of VMs, the identifying based at least in part on the third VM operating on a host having a highest load among hosts in the plurality of hosts that have a load below a second threshold, the second threshold below the first threshold and indicative of running the third VM in a low power state; and

send the third VM an indication to conduct the second remote presentation session.

17. The system of claim 11 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

determine that a state of a second VM of the plurality of VMs is incorrect based on a load of the second VM; and

change the state of the second VM.

18. The system of claim 17 , wherein the memory further bears instructions that, upon execution by the processor, cause the system at least to:

receive an indication of the load of the second VM from a virtual machine monitor (VMM) that manages the execution of the second VM.

19. The system of claim 18 , wherein the VMM monitors the load of the second VM based on a system call made from the second VM to a host on which the second VM and the VMM execute.

20. A computer-readable storage medium, excluding propagation media, bearing computer-executable instructions that, upon execution by a computer, cause the computer to perform operations comprising:

receiving a request to connect a client to a remote presentation session;

making a first determination that no VM of the plurality of VMs is both set to accept remote presentation session connections and has a load less than a first threshold indicative of a level of computer resource usage above which remote presentation sessions should not be added;

identifying, in response to the first determination, a subset of the plurality of VMs, each VM of the subset being set to a drain state indicative of allowing the VM of the subset to enter a sleep state;

identifying a first VM in the subset having a highest load of VMs in the subset;

setting the first VM to accept remote presentation session connections;

sending the first VM an indication to conduct the remote presentation session with the client;

receiving a request to connect a second remote presentation session; and

sending a second VM an indication to conduct the second remote presentation session upon determining that no VM of the plurality of VMs has a load below the first threshold and determining that a second VM of the plurality of VMs has a lowest load above the first threshold.