# US 2014/0207425 A1 · System, Method and Apparatus for Adaptive Virtualization > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0207425 A1 - Document ID: US20140207425A1 - Status: Published application - Kind: A1 - Application number: 13/744,572 - Filed: 2013-01-18 - Published: 2014-07-24 - Earliest claimed: 2013-01-18 - Inventors: Michael Yeung; Thomas C. Bootland; Tom Gray - Web page: https://patentgrep.com/patent/US20140207425A1 ## CPC - G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model - G06F 9/5072: {Grid computing} - G06F 9/5077: {Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455)} - G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A method for a computing apparatus, of adopting virtualization resource requirements is provided. Performance indicators at least in part related to a virtualized application are received and a vendor model is updated based on the received performance indicators. The method also allows for generating, based on the updated vendor model, updated resource requirement estimates for the virtualized application. The updated resource requirements serve as a basis for sending, to a cloud service provider, a capacity change request based on the updated resource requirement estimates. The timing for generating and sending updates is managed by an adaptation module and is based on an update policy. The update policy itself can be based on schedules that reflect estimated capacity changes for the virtualized application. The updating can be based on a learning module which can use Kalman filters and other techniques for learning. ## Claims ### Claim 1 A method, for a computing apparatus, of adapting virtualization resource requirements, said method comprising: receiving performance indicators; updating a vendor model based on said received performance indicators. ### Claim 2 (depends on 1) The method of claim 1 further comprising: generating, based on said updated vendor model, updated resource requirement estimates for a virtualized application; sending, to a cloud service provider, a capacity change request based on the updated resource requirement estimates. ### Claim 3 (depends on 2) The method of claim 2, wherein timing of said generating and sending is based on an update policy. ### Claim 4 (depends on 3) The method of claim 3, wherein said update policy is based on at least one schedule that reflects estimated capacity changes for a virtualized application. ### Claim 5 (depends on 4) The method of claim 4, wherein said schedule is developed based on historical data. ### Claim 6 (depends on 1) The method of claim 1, wherein at least one of said performance indicators is received from a virtualized application. ### Claim 7 (depends on 6) The method of claim 6, wherein said at least one of said performance indicators is based on one of a quality of service provided to users of said virtualized application and internal metrics on the operation of said virtualized application. ### Claim 8 (depends on 1) The method of claim 1, wherein said updating is based on a Kalman filter. ### Claim 9 (depends on 1) The method of claim 1 further comprising transforming said performance indicators into modification indicators. ### Claim 10 (depends on 9) The method of claim 9, wherein transforming is accomplished using fuzzy averaging techniques. ### Claim 11 (depends on 10) The method of claim 10, wherein said transforming further comprises providing performance estimates based on said performance indicators based on a predetermined value, wherein said fuzzy averaging techniques are performed based on said performance estimates. ### Claim 12 (depends on 11) The method of claim 11, wherein each said performance estimate is associated with one or more related resources. ### Claim 13 (depends on 12) The method of claim 12, wherein each said performance estimate associated with one or more resources has a corresponding confidence level associated with said one of more associated resources. ### Claim 14 (depends on 3) The method of claim 3, wherein said update policy further includes determining whether said updated vendor model varies significantly from said vendor model prior to updating. ### Claim 15 (depends on 2) The method of claim 2, wherein said updating also generates a request for scheduling said generating and sending. ### Claim 16 (depends on 2) The method of claim 2, wherein said sending is based on a signal by an operator model. ### Claim 17 (depends on 1) The method of claim 1, wherein said updating is further based on learning policies. ### Claim 18 (depends on 17) The method of claim 17, wherein said learning policies are based on periods where a capacity needed approaches maximum capacity estimated by said vendor model. ### Claim 19 A computing apparatus for adaptive virtualization comprising: an input device for receiving performance indicators; and a processor configured for updating a vendor model based on said received performance indicators. ### Claim 20 (depends on 19) The computing apparatus of claim 19, wherein said processor is further configured for generating, based on said updated vendor model, updated resource requirement estimates for a virtualized system, said computing apparatus further comprising an output device for sending, to a cloud service provider, a capacity change request based on said updated resource requirement estimates.