Distributed policy-based provisioning and enforcement for quality of service

• US 9,887,924 B2
• Filed: 08/26/2013
• Issued: 02/06/2018
• Est. Priority Date: 08/26/2013
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Earl, William, Karamanolis, Christos
• Primary Examiner(s): Cheema, Umar
• Assistant Examiner(s): Mekonen, Tesfu
• Application Number:
  US14/010,247
• Publication Number:
  US 20150058475A1
• Time in Patent Office: 1,625 Days
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 3/0605 by facilitating the interaction with a user or administrator
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 9/4411 Configuring for operating with peripheral devices; Loading of device drivers
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 47/24 Traffic characterised by specific attributes, e.g. priority or QoS
• CPC Class Codes: H04L 47/2483 involving identification of individual flows
• CPC Class Codes: H04L 47/741 Holding a request until resources become available
• CPC Class Codes: H04L 47/822 Collecting or measuring resource availability data
• CPC Class Codes: H04L 47/83 based on usage prediction

Abstract

Embodiments of the disclosure provide techniques for measuring congestion and controlling quality of service to a shared resource. A module that interfaces with the shared resource monitors the usage of the shared resource by accessing clients. Upon detecting that the rate of usage of the shared resource has exceeded a maximum rate supported by the shared resource, the module determines and transmits a congestion metric to clients that are currently attempting to access the shared resource. Clients, in turn determine a delay period based on the congestion metric prior to attempting another access of the shared resource.

Claims

1. A computer-implemented method for providing resource usage feedback to a plurality of clients having access to a shared storage resource, the method comprising:

monitoring a rate of usage of the shared storage resource by at least a portion of the clients, wherein a minimum usage rate for the shared storage resource is reserved for each client;

upon detecting that the rate of usage of the shared storage resource as monitored has exceeded a maximum rate supported by the shared storage resource, determining a congestion metric for each of the clients that are currently attempting to access the shared storage resource, wherein the congestion metric for each client is based on usage by each client of the shared storage resource; and

transmitting the corresponding determined congestion metric to each of the clients that are currently attempting to access the shared storage resource for each of the clients to introduce no delay or non-zero delay based on the corresponding determined congestion metric prior to attempting another access of the shared storage resource, wherein all clients that have exceeded their reserved minimum usage rate for the shared storage resource are transmitted a congestion metric having a non-zero value that results in a delay period being calculated by the client and all clients that have not exceeded their reserved minimum usage rate for the shared storage resource are transmitted a congestion metric having a zero value that results in no delay period being calculated by the client.

2. The method of claim 1,

wherein the shared storage resource is a local storage drive housed in a host computer belonging to a cluster of host computers that supports the execution of the plurality of clients and

wherein each host computer in the cluster contributes a corresponding local storage drive to form a pool of shared storage resources accessible as a single virtual storage device.

3. The method of claim 2,

wherein at least one of the clients is performing I/O operations on a virtual disk, and

wherein at least one section of the virtual disk is backed by a portion of the shared storage resource and at least one other section of the virtual disk is backed by a different local storage disk housed in a different host computer belonging to the cluster.

4. The method of claim 3, wherein the virtual disk is organized as a RAID 0 stripe set and the section of the virtual disk backed by a portion of the shared storage resource and the at least one other section of the virtual disk backed by a different local storage disk both represent a different stripe in the stripe set.

5. The method of claim 3,

wherein a second client is performing I/O operations on a second virtual disk, and

wherein at least one portion of the second virtual disk is backed by a second portion of the shared storage resource and at least one other portion of the second virtual disk is backed by a different local disk housed in a different host computer belonging to the cluster.

6. The method of claim 1, wherein the clients each comprise a virtual machine running in one of the host computers in the cluster.

7. A non-transitory computer readable storage medium storing instructions, which, when executed on a processor, performs an operation for providing resource usage feedback to a plurality of clients having access to a shared storage resource, the operation comprising:

monitoring a rate of usage of the shared storage resource by at least a portion of the clients, wherein a minimum usage rate for the shared storage resource is reserved for each client;

upon detecting that the rate of usage of the shared storage resource as monitored has exceeded a maximum rate supported by the shared storage resource, determining a congestion metric for at least a portion each of the clients that are currently attempting to access the shared storage resource, wherein the congestion metric for each client is based on usage by each client of the shared storage resource; and

transmitting the corresponding determined congestion metric to each of the clients that are currently attempting to access the shared storage resource for each of the clients to introduce no delay or non-zero delay based on the corresponding determined congestion metric prior to attempting another access of the shared storage resource, wherein all clients that have exceeded their reserved minimum usage rate for the shared storage resource are transmitted a congestion metric having a non-zero value that results in a delay period being calculated by the client and all clients that have not exceeded their reserved minimum usage rate for the shared storage resource are transmitted a congestion metric having a zero value that results in no delay period being calculated by the client.

8. The non-transitory computer readable storage medium of claim 7,

wherein the shared storage resource is a local storage drive housed in a host computer belonging to a cluster of host computers that supports the execution of the plurality of clients and

wherein each host computer in the cluster contributes a corresponding local storage drive to form a pool of shared storage resources accessible as a single virtual storage device.

9. The non-transitory computer readable storage medium of claim 8,

wherein at least one of the clients is performing I/O operations on a virtual disk, and

wherein at least one section of the virtual disk is backed by a portion of the shared storage resource and at least one other section of the virtual disk is backed by a different local storage disk housed in a different host computer belonging to the cluster.

10. The non-transitory computer readable storage medium of claim 9, wherein the virtual disk is organized as a RAID 0 stripe set and the section of the virtual disk backed by a portion of the shared storage resource the at least one other section of the virtual disk backed by a different local storage disk both represent a different stripe in the stripe set.

11. The non-transitory computer readable storage medium of claim 9,

wherein a second client is performing I/O operations on a second virtual disk, and

wherein at least one portion of the second virtual disk is backed by a second portion of the shared storage resource and at least one other portion of the second virtual disk is backed by a different local disk housed in a different host computer belonging to the cluster.

12. The non-transitory computer readable storage medium of claim 7, wherein the clients each comprise a virtual machine running in one of the host computers in the cluster.

13. A computer system, comprising:

a processor and a memory hosting a module, which, when executed on the processor, performs an operation for providing resource usage feedback to a plurality of clients having access to a shared storage resource, the operation comprising;

monitoring a rate of usage of the shared storage resource by at least a portion of the clients, wherein a minimum usage rate for the shared storage resource is reserved for each client;

upon detecting that the rate of usage of the shared storage resource as monitored has exceeded a maximum rate supported by the shared storage resource, determining a congestion metric for each of the clients that are currently attempting to access the shared storage resource, wherein the congestion metric for each client is based on usage by each client of the shared storage resource; and

transmitting the corresponding determined congestion metric to each of the clients that are currently attempting to access the shared storage resource for each of the clients to introduce no delay or non-zero delay based on the corresponding determined congestion metric prior to attempting another access of the shared storage resource, wherein all clients that have exceeded their reserved minimum usage rate for the shared storage resource are transmitted a congestion metric having a non-zero value that results in a delay period being calculated by the client and all clients that have not exceeded their reserved minimum usage rate for the shared storage resource are transmitted a congestion metric having a zero value that results in no delay period being calculated by the client.

14. The computer system of claim 13,

wherein the shared storage resource is a local storage drive housed in a host computer belonging to a cluster of host computers that supports the execution of the plurality of clients and

wherein each host computer in the cluster contributes a corresponding local storage drive to form a pool of shared storage resources accessible as a single virtual storage device.

15. The computer system of claim 14,

wherein at least one of the clients is performing I/O operations on a virtual disk, and

wherein at least one section of the virtual disk is backed by a portion of the shared storage resource and at least one other section of the virtual disk is backed by a different local storage disk housed in a different host computer belonging to the cluster.

16. The computer system of claim 15, wherein the virtual disk is organized as a RAID 0 stripe set and the section of the virtual disk backed by a portion of the shared storage resource and the at least one other section of the virtual disk backed by a different local storage disk both represent a different stripe in the stripe set.

17. The computer system of claim 15,

wherein a second client is performing I/O operations on a second virtual disk, and

wherein at least one portion of the second virtual disk is backed by a second portion of the shared storage resource and at least one other portion of the second virtual disk is backed by a different local disk housed in a different host computer belonging to the cluster.