Methods and systems that allocate cost of cluster resources in virtual data centers

• US 9,747,136 B2
• Filed: 02/11/2015
• Issued: 08/29/2017
• Est. Priority Date: 12/09/2014
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Sathyamurthy, Dattathreya, Gaurav, Kumar, Masrani, Piyush Bharat, Kumar Pannem, Hemanth, Pulavarthi, Chintesh
• Primary Examiner(s): GHAFFARI, ABU Z
• Application Number:
  US14/619,071
• Publication Number:
  US 20160162338A1
• Time in Patent Office: 930 Days
• Field of Search: None
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/455 Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5072 Grid computing
• 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: G06Q 30/04 Billing or invoicing
• CPC Class Codes: H04L 47/821 Prioritising resource allocation or reservation requests
• CPC Class Codes: H04L 47/823 Prediction of resource usage

Abstract

This disclosure presents computational systems and methods that allocate cost of resources of a cluster of server computers used by virtual machines in a virtual data center. In one aspect, a fair unit rate is computed based on the larger of a measured average utilization or an expected utilization of a cluster resource of server computers within a physical data center by virtual machines. The fair unit rate is a cost per unit of resource used over a period of time and is used to compute an allocated cost of the virtual machine usage of the cluster resource.

Claims

1. A method that uses a programmable computer programmed to allocate cost of cluster resources of a cluster of server computers, the method comprising:

computing a measured average utilization of a cluster resource of the cluster of server computers within a physical data center by a number of virtual machines;

computing an expected utilization of the cluster resource based on a total capacity of the cluster resource and a reserved capacity of the cluster resource;

computing a fair unit rate based on a fully loaded cost of the cluster resource and maximum of one of the expected utilization and the measured average utilization;

computing an allocated cost of the utilization of the cluster resource by the virtual machines based on the fair unit rate and each virtual machines utilization of the cluster resource;

computing an unallocated cost of the cluster resource based on the measured average utilization and the expected utilization of the cluster resource; and

re-provisioning the cluster resource for the virtual machines when the unallocated cost is greater than a threshold in order to optimize use of the cluster resource within the physical data center.

2. The method of claim 1 wherein computing the unallocated cost of the cluster resource comprises computing the unallocated cost of the cluster resource when the measured average utilization is less than the expected utilization based on the fully loaded cost of the cluster resource, expected utilization and difference between the expected utilization and the measured average utilization.

3. The method of claim 1 wherein computing the measured average utilization further comprises:

measuring utilization of the cluster resource by each of the virtual machines over the period of time;

computing an average measured utilization of the cluster resource by each of the virtual machines over the period of time; and

summing the average measured utilization to generate the measured average utilization of the cluster resource.

4. The method of claim 1 wherein the expected utilization is the difference between the total capacity of the cluster resource and a reserved capacity of cluster resource.

5. The method of claim 1 further comprises repeating the operations of computing the measured average utilization, computing the expected utilization, computing the fair unit rate, computing the allocated cost for each cluster resource of the cluster of server computers and summing the allocated cost to generate a total allocated cost of cluster resources.

6. A computer system comprising:

one or more processors;

one or more data-storage devices; and

a routine stored in the data-storage devices and executed using the one or more processors, the routine

computing a measured average utilization of a cluster resource of a cluster of server computers within a physical data center by a number of virtual machines;

computing an expected utilization of the cluster resource based on a total capacity of the cluster resource and a reserved capacity of the cluster resource;

computing a fair unit rate based on a fully loaded cost of the cluster resource and maximum of one of the expected utilization and the measured average utilization;

computing an allocated cost of the utilization of the cluster resource by the virtual machines based on the fair unit rate and each virtual machines utilization of the cluster resource;

computing an unallocated cost of the cluster resource based on the measured average utilization and the expected utilization of the cluster resource; and

re-provisioning the cluster resource for the virtual machines when the unallocated cost is greater than a threshold in order to optimize use of the cluster resource within the physical data center.

7. The computer system of claim 6 wherein computing the unallocated cost of the cluster resource comprises computing the unallocated cost of the cluster resource when the measured average utilization is less than the expected utilization based on the fully loaded cost of the cluster resource, expected utilization and difference between the expected utilization and the measured average utilization.

8. The computer system of claim 6 wherein computing the measured average utilization further comprises:

measuring utilization of the cluster resource by each of the virtual machines over the period of time;

computing an average measured utilization of the cluster resource by each of the virtual machines over the period of time; and

summing the average measured utilization to generate the measured average utilization of the cluster resource.

9. The computer system of claim 6 wherein the expected utilization is the difference between the total capacity of the cluster resource and a reserved capacity of cluster resource.

10. The computer system of claim 6 further comprises repeating the operations of computing the measured average utilization, computing the expected utilization, computing the fair unit rate, computing the allocated cost for each cluster resource of the cluster of server computers and summing the allocated cost to generate a total allocated cost of cluster resources.

11. A non-transitory computer-readable medium encoded with machine-readable instructions that implement a method carried out by one or more processors of a computer system to perform the operations of computing a measured average utilization of a cluster resource of a cluster of server computers within a physical data center by a number of virtual machines;

computing an expected utilization of the cluster resource based on a total capacity of the cluster resource and a reserved capacity of the cluster resource;

computing a fair unit rate based on a fully loaded cost of the cluster resource and maximum of one of the expected utilization and the measured average utilization;

computing an allocated cost of the utilization of the cluster resource by the virtual machines based on the fair unit rate and each virtual machines utilization of the cluster resource;

computing an unallocated cost of the cluster resource based on the measured average utilization and the expected utilization of the cluster resource; and

re-provisioning the cluster resource for the virtual machines when the unallocated cost is greater than a threshold in order to optimize use of the cluster resource within the physical data center.

12. The medium of claim 11 wherein computing the unallocated cost of the cluster resource comprises computing the unallocated cost of the cluster resource when the measured average utilization is less than the expected utilization based on the fully loaded cost of the cluster resource, expected utilization and difference between the expected utilization and the measured average utilization.

13. The medium of claim 11 wherein computing the measured average utilization further comprises:

measuring utilization of the cluster resource by each of the virtual machines over the period of time;

computing an average measured utilization of the cluster resource by each of the virtual machines over the period of time; and

summing the average measured utilization to generate the measured average utilization of the cluster resource.

14. The medium of claim 11 wherein the expected utilization is the difference between the total capacity of the cluster resource and a reserved capacity of cluster resource.

15. The medium of claim 11 further comprises repeating the operations of computing the measured average utilization, computing the expected utilization, computing the fair unit rate, computing the allocated cost for each cluster resource of the cluster of server computers and summing the allocated cost to generate a total allocated cost of cluster resources.