METHODS AND APPARATUS TO MANAGE COMPUTE RESOURCES IN A HYPERCONVERGED INFRASTRUCTURE COMPUTING ENVIRONMENT

• US 20190034237A1
• Filed: 09/18/2018
• Published: 01/31/2019
• Est. Priority Date: 02/25/2017
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Siddappa, Gireesha Udagani, Nayak, Samdeep, Kottapalli, Ravi Kumar Reddy, Hemige, Srinivas Sampatkumar, Verma, Shubham
• Granted Patent: US 10,684,893 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 2009/45587 Isolation or security of virtual machine instances
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: 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
• CPC Class Codes: H04L 67/1012 based on compliance of requirements or conditions with available server resources
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests

Abstract

Methods, apparatus, systems and articles of manufacture are disclosed for managing compute resources in a hyperconverged infrastructure computing environment. An example apparatus includes a resource manager table populator to populate a resource management table with information identifying workloads and a workload categorizer to categorize the workloads identified in the resource management table as needy workloads or offering workloads. The disclosed apparatus further include a workload selector to select an offering workload to lend resources to a needy workload, and a resource configurer. The resource configurer to cause a host included in the workload to spin a virtual machine or create a container, and assign the virtual machine or container to support the needy workload.

Claims

1. An apparatus to manage workload resources in a hyperconverged infrastructure computing environment comprising:

a resource manager table populator to populate a resource management table with information identifying workloads;

a workload categorizer to categorize the workloads identified in the resource management table as needy workloads or offering workloads;

a workload selector to select an offering workload to lend resources to a needy workload; and

a resource configurer to;

cause a host included in the workload to spin a virtual machine or create a container; and

assign the virtual machine or container to support the needy workload.

2. The apparatus defined in claim 1, wherein the resource configurer causes the host to spin the virtual machine by causing a virtual machine spinner to generate an instruction to the host, the instruction to cause the host to generate the virtual machine.

3. The apparatus defined in claim 1, wherein the workload categorizer is to categorize a first workload as a first needy workload when the first workload is using a first threshold amount of a first amount of first resources assigned to support the first workload, and the workload categorizer is to categorize a second workload as a second offering workload when the second workload has more than a second threshold amount of a second amount of second resources assigned to support the second workload.

4. An apparatus as defined in claim 1, wherein the virtual machine is a first virtual machine, and the resource configurer causes the host to spin the first virtual machine when the host is implemented with a second virtual machine.

5. An apparatus as defined in claim 1, wherein the container is a first container, and the resource configurer causes the host to spin the first container when the host is implemented with a second container.

6. An apparatus as defined in claim 1, wherein the resource manager table generator/populator populates the resource management table with workload information obtained by polling network resources, the workload information to include workload identification information that distinguishes the workloads from one another, and host identifiers identifying hosts assigned to support a workload, a first amount of compute resource currently being utilized and a second amount of compute resource assigned to support the workload.

7. An apparatus as defined in claim 1, wherein the workload selector is to identify a first offering workload of a plurality of offering workloads to lend resources to the needy workload based on first characteristics of the first offering workload being similar to second characteristics of the needy workload, the first and second characteristics to include at least of an operating speed and a security configuration.

8. A non-transitory computing readable medium comprising instructions that, when executed, cause a machine to at least:

track respective percentages of available resources used by a plurality of respective workloads, the plurality of workloads including a first workload and a second workload, the workloads operating in a hyperconverged infrastructure computing environment;

based on the respective percentages, assign one of a plurality of categories to the respective workloads;

based on the respective categories assigned to the respective first and second workloads, cause a host included in the first workload to spin a virtual machine or create a container; and

assign the virtual machine or the container to the second workload.

9. The non-transitory computing readable medium defined in claim 8, wherein the plurality of categories indicate an ability of the respective workloads to lend resources to another of the workloads.

10. The non-transitory computing readable medium defined in claim 8, wherein the instructions, when executed, cause the machine to:

store the respective percentages of available resources and the respective workload identifiers in a table; and

access the table when assigning the one of the plurality of categories to the respective workloads.

11. The non-transitory computing readable medium defined in claim 8, wherein the plurality of categories include a first category and a second category, the first category to be assigned to workloads that are using at least a first threshold amount of allocated resources, and the second category to be assigned to workloads that are not using at least a second amount of allocated resources.

12. The non-transitory computing readable medium defined in claim 8, wherein the virtual machine is a first virtual machine, and the instructions cause the machine to cause the host included in the first workload to spin the first virtual machine when the host is implemented with a second virtual machine.

13. The non-transitory computing readable medium defined in claim 8, wherein the container is a first container, and the instructions cause the machine to cause the host included in the first workload to spin the container when the host is implemented with a second container.

14. The non-transitory computing readable medium defined in claim 9, wherein the instructions further cause the machine to select the host of the first workload to spin the virtual machine or the container based, at least in part, on first characteristics of the first workload being similar to second characteristics of the second workload, the first and second characteristics to include at least of an operating speed and a security configuration.

15. A method to manage compute resources in a hyperconverged infrastructure computing environment, the method comprising:

tracking respective percentages of available resources used by a plurality of respective workloads, the plurality of workloads including a first workload and a second workload;

based on the respective percentages, assigning one of a plurality of categories to the respective workloads;

based on the respective categories assigned to the respective first and second workloads, causing a host included in the first workload to spin a virtual machine or create a container; and

assigning the virtual machine or the container to support the second workload.

16. The method defined in claim 15, wherein the plurality of categories indicate an ability of the respective workloads to lend resources to another of the workloads.

17. The method defined in claim 15, further including

storing the respective percentages of available resources and the respective workload identifiers in a table;

and

accessing the table when assigning the one of the plurality of categories to the respective workloads.

18. The method defined in claim 15, wherein the plurality of categories include a first category and a second category, the first category to be assigned to workloads that are using at least a first threshold amount of allocated resources, and the second category to be assigned to workloads that are not using at least a second amount of allocated resources.

19. The method defined in claim 15, wherein the virtual machine is a first virtual machine, and the method further includes causing the machine to cause the host included in the first workload to spin the first virtual machine when the host is implemented with a second virtual machine.

20. The method defined in claim 15, wherein the container is a first container, and the method further includes causing the host included in the first workload to spin the container when the host is implemented with a second container.