US 12,153,951 B2
System and method for managing workload of an application in a cloud computing environment
Roopak Parikh, San Jose, CA (US); Madhura Maskasky, Los Altos, CA (US); Pushkar Acharya, San Jose, CA (US); Mayuresh Kulakarni, Pune (IN); Ashutosh Tiwari, Pune (IN); Anirudh Pokala, Telangana (IN); Omkar Deshpande, Maharashtra (IN); and Shubham Agarwal, Rajasthan (IN)
Assigned to Platform9, Inc., San Jose, CA (US)
Filed by Platform9, Inc., San Jose, CA (US)
Filed on Feb. 1, 2024, as Appl. No. 18/429,946.
Prior Publication US 2024/0220307 A1, Jul. 4, 2024
Int. Cl. G06F 9/455 (2018.01)
CPC G06F 9/45558 (2013.01) [G06F 2009/45583 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method for managing workload of an application in a cloud computing environment, the method comprising:
receiving, by a processor, cloud service data from an existing cloud infrastructure, wherein the existing cloud infrastructure (ECI) comprises a storage system, one or more bare metal servers, and a network interface, a storage interface, a cloud control manager, and one or more virtual machines (VMs);
configuring, by the processor, the cloud control manager to connect an Elastic Machine Pool Infrastructure (EMPI) to the existing cloud infrastructure, wherein the cloud control manager is configured to communicate with the network interface and the storage interface to connect the EMPI to the ECI, and wherein the EMPI comprises an orchestrator;
receiving, by the processor, the workload from the application running on the cloud computing environment;
creating, by the processor, one or more Elastic Virtual Machines (EVMs) in an Elastic Machine Pool (EMP), using the orchestrator, based on the received workload;
allocating, by the processor, the workload to at least one of a VM and an EVM based on at least one of an EMP profile of the application, status of the VM, status of the EVM, and workload characteristics, wherein the EVM is hosted on one of the one or more bare metal servers; and
managing, by the processor, the one or more bare metal servers and the one or more EVMs based on the workload characteristics, the status of the VM, node management data, and the status of the EVM, wherein the one or more bare metal servers and the one or more EVMs are managed using a machine learning algorithm, and wherein the machine learning algorithm is trained using the node management data.