US 11,900,171 B2
Cloud computing capacity management system using automated fine-grained admission control
Gurpreet Virdi, Redmond, WA (US); Fernando Gonzalez Todisco, Bellevue, WA (US); Karthikeyan Subramanian, Redmond, WA (US); Sanjay Ramanujan, Issaquah, WA (US); Sorin Iftimie, Sammamish, WA (US); Xing wen Wang, Redmond, WA (US); Thomas Moscibroda, Bellevue, WA (US); Yunus Mohammed, Redmond, WA (US); Vi Lam Nguyen, Redmond, WA (US); and Rostislav Sudakov, Redmond, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Feb. 2, 2021, as Appl. No. 17/165,904.
Prior Publication US 2022/0245001 A1, Aug. 4, 2022
Int. Cl. G06F 9/50 (2006.01); G06F 11/34 (2006.01)
CPC G06F 9/5072 (2013.01) [G06F 9/505 (2013.01); G06F 9/5077 (2013.01); G06F 11/3433 (2013.01); G06F 2209/503 (2013.01); G06F 2209/505 (2013.01); G06F 2209/5019 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A cloud computing capacity management system, comprising:
one or more processors;
a prediction engine configured to generate one or more capacity signals;
a fine-grained admission control layer that is executable by the one or more processors to ingest the capacity signals and create a capacity mitigation policy, based at least in part on the capacity signals, to protect available capacity of a cloud computing system, wherein the capacity mitigation policy is directed to a deployment grain, the deployment grain indicating a subscriber identifier and a resource identifier associated with a set of one or more users of the cloud computing system;
a policy engine that is executable by the one or more processors to control how the capacity mitigation policy is applied to the set of one or more users of the deployment grain in order to cause one or more capacity mitigation actions to be performed when capacity shortages are predicted and to undo the one or more capacity mitigation actions when the capacity shortages are no longer predicted;
an enforcement layer that is executable by the one or more processors to handle incoming resource requests from the set of one or more users associated with the deployment grain and to enforce resource limits based on the capacity mitigation policy as applied by the policy engine; and
a manual override lever for managing extraneous situations outside of intelligence provided by the prediction engine to update the capacity mitigation policy.