Load Balancing By Endpoints
Abstract
A mechanism is provided for In a cloud computing infrastructure, a mechanism is provided for balancing client sessions across virtual machines such that the number of virtual machines is efficiently managed. In some embodiments, the total number of virtual machines is minimized to reduce power consumption, cooling, and other cost drivers, while assigning users across the sessions. In one embodiment, the sessions in a virtual machine with low activity are migrated to a virtual machine with higher session rates to allow for the shutdown of the low usage virtual machines. In another embodiment, new user sessions are assigned according to a minimum performance standard.
Claims
exact text as granted — not AI-modified1 . A method for managing a virtualized computing infrastructure, the method comprising:
receiving a request for a remote desktop session from one of a plurality of users; determining the number of remote desktop sessions hosted by each of a plurality of virtual machines; selecting a virtual machine to host the requested remote desktop session in order to increase the number of remote desktop sessions hosted by the virtual machines that currently host a plurality of remote desktop sessions; instantiating the requested remote desktop session on the selected virtual machine.
2 . The method of claim 1 , wherein the selected virtual machine is determined by balancing loads among virtual machines such that new user sessions are hosted on computing devices to substantially minimize the total number of virtual machines.
3 . The method of claim 2 , wherein the selected virtual machine is the virtual machine with the highest number of hosted remote desktop sessions.
4 . The method of claim 2 , wherein the selected virtual machine is the virtual machine with fewer than the highest number of hosted remote desktop sessions and more than the lowest number of hosted remote desktop sessions.
5 . The method of claim 4 , further comprising classifying each virtual machine as having one of a low, medium, and high load level, and assigning new user sessions to the computing devices in the order of medium, high, and low.
6 . The method of claim 1 , wherein said selecting further comprises assigning the remote desktop session to an endpoint according to a random selection method.
7 . The method of claim 1 , wherein said selecting further comprises assigning the remote desktop session to an endpoint according to one of a round robin, queuing, and session caching selection method.
8 . The method of claim 1 , wherein the predetermined load balancing metric is determined according to a minimum performance standard.
9 . A computing system comprising:
a computing device comprising at least one processor; a memory communicatively coupled to said processor when said system is operational; said memory having stored therein computer instructions that upon execution by the at least one processor cause: instantiating one or more remote desktop configurations that makes available, to a plurality of users via a remote network connection, remote desktop configurations each comprising an operating environment and software applications to be included in the operating environment; receiving a request for a remote desktop session from one of the plurality of users, the remote desktop session conforming to one of the one or more remote desktop configurations; and instantiating the requested remote desktop session and loading data and applications associated with the remote desktop session on a computing device selected in accordance with a predetermined load balancing metric.
10 . The computing system of claim 9 , further comprising migrating the data and applications to another computing device in accordance with the predetermined load balancing metric.
11 . The computing system of claim 9 , wherein the one or more remote desktop configurations each conform to a user configuration template for scalably propagating the one or more remote desktop configurations in a virtualized computing environment.
12 . The computing system of claim 9 , wherein the metric is determined by balancing loads among computing devices such that new user sessions are hosted on computing devices to substantially minimize the total number of hosting computing devices.
13 . The computing system of claim 12 , wherein the new user sessions are hosted on an available computing device with the highest number of hosted sessions.
14 . The computing system of claim 12 , further comprising classifying each computing device as having one of a low, medium, and high load level, and assigning new user sessions to the computing devices in the following order: medium, high, and low.
15 . The computing system of claim 14 , wherein user sessions hosted on a computing device classified as low are migrated to a computing device classified as high to allow for a shutdown of the computing device classified as low.
16 . The computing system of claim 9 , wherein said selecting further comprises assigning the remote desktop session to an endpoint according to one of a random, round robin, queuing, and session caching selection method.
17 . A computer readable storage medium storing thereon computer executable instructions for providing access to a remote user session in a computing environment, the computer readable storage medium comprising:
instructions for saving, at a first computing data center, a user remote desktop configuration, the user remote desktop configuration including saved state information for an operating environment and software applications executing in the operating environment, the first computing data center configured to provide at least computation and storage services; instructions for determining a probable location from which a user will access the user remote desktop configuration; and instructions for based on said determining, migrating the saved user remote desktop configuration to a second computing data center and making available, via a remote network connection, the saved user remote desktop configuration to the user from the second computing data center.
18 . The computer readable storage medium of claim 17 , wherein the metric is determined by balancing loads among computing devices such that new user sessions are hosted on computing devices to substantially minimize the total number of hosting computing devices.
19 . The computer readable storage medium of claim 18 , wherein the new user sessions are hosted on an available computing device with the highest number of hosted sessions.
20 . The computer readable storage medium of claim 18 , further comprising instructions for classifying each computing device as having one of a low, medium, and high load level, and instructions for assigning new user sessions to the computing devices in the following order: medium, high, and low.Join the waitlist — get patent alerts
Track US2013067469A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.