Distributed policy-based provisioning and enforcement for quality of service

Patent metadata

Field	Value
Publication number	US-10855602-B2
Application number	US-201815889196-A
Country	US
Kind code	B2
Filing date	Feb 5, 2018
Priority date	Aug 26, 2013
Publication date	Dec 1, 2020
Grant date	Dec 1, 2020

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.

First claim

We claim:

1. A computer-implemented method for handling a storage input/output (I/O) operation originating from a virtual machine (VM) and destined for a virtual disk, the VM running on virtualization software of a first host that is part of a cluster of host computers that includes the first host and a second host, each host having its own local physical storage resources, the virtualization software of the first host containing a first driver, the method comprising:
receiving, by the first driver, the I/O operation;
retrieving from the I/O operation an identifier for a virtual disk object representing the virtual disk, the virtual disk object being composed of virtual component objects, the virtual component objects being mappable to any available local physical storage resources throughout the cluster of host computers;
determining, using the identifier, that a second host is a virtual disk coordinator host of the virtual disk object, by accessing a database of the first host,
wherein the virtual disk coordinator host controls access to the virtual disk object;
performing one of:
(1) transmitting, by the first driver, the I/O operation to a second driver of the second virtualization software of the second host, or
(2) updating the database to set the first host to be the virtual disk coordinator host of the virtual disk object; and
executing the I/O operation by the virtual disk coordinator host by communicating with hosts that are the coordinators for the virtual component objects of the virtual disk object.

2. The method of claim 1 , wherein the first host implements a hypervisor, and the hypervisor comprises the first driver.

3. The method of claim 1 , wherein the performing comprises transmitting, by the first driver, the I/O operation to the second driver of the virtual disk coordinator host.

4. The method of claim 1 , wherein the performing comprises updating the database to set the first host to be the virtual disk coordinator host of the virtual disk object.

5. The method of claim 1 , wherein the executing of the I/O operation by communicating with hosts that are the coordinators for the virtual component objects of the virtual disk object comprises:
identifying, by the virtual disk coordinator host, virtual component objects of the virtual disk that are subject to the I/O operation, wherein a virtual component object encompasses a data partition of the virtual disk;
determining a component object coordinator host for each of the virtual component objects, wherein the component object coordinator host controls access to an associated virtual component object;
transmitting, by the virtual disk coordinator host, the I/O operation to each of the component object coordinator hosts; and
performing, by each of the component object coordinator hosts, the I/O operation in a local storage resource backing the virtual component object.

6. The method of claim 5 , wherein each host computer in the cluster contributes at least a portion of a corresponding local storage disk to form a pool of shared resources accessible as the virtual disk.

7. The method of claim 6 , wherein
at least one section of the virtual disk is backed by a local storage disk housed in the first host, 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.

8. The method of claim 7 , wherein the virtual disk is organized as a RAID 0 stripe set, and wherein (1) the section of the virtual disk backed by the local storage disk housed in the first host, and (2) the at least one other section of the virtual disk backed by the different local storage disk represent different stripes in the stripe set.

9. A non-transitory computer readable medium comprising instructions to be executed in a processor of a computer system, and the instructions when executed in the processor cause the computer system to carry out a method of handling a storage input/output (I/O) operation originating from a virtual machine (VM) and destined for a virtual disk, the VM running on virtualization software of a first host that is part of a cluster of host computers that includes the first host and a second host, each host having its own local physical storage resources, the virtualization software of the first host containing a first driver, said method comprising:
receiving, by the first driver, the I/O operation;
retrieving from the I/O operation an identifier for a virtual disk object representing the virtual disk, the virtual disk object being composed of virtual component objects, the virtual component objects being mappable to any available local physical storage resources throughout the cluster of host computers;
determining, using the identifier, that a second host is a virtual disk coordinator host of the virtual disk object by accessing a database of the first host, wherein the virtual disk coordinator host controls access to the virtual disk object;
performing one of:
(1) transmitting, by the first driver, the I/O operation to a second driver of the second virtualization software of the second host virtual disk coordinator host, or
(2) updating the database to set the first host to be the virtual disk coordinator host of the virtual disk object; and
executing the I/O operation by the virtual disk coordinator host by communicating with hosts that are the coordinators for the virtual component objects of the virtual disk object.

10. The non-transitory computer readable medium of claim 9 , wherein the first host implements a hypervisor, and the hypervisor comprises the first driver.

11. The non-transitory computer readable medium of claim 9 , wherein the performing comprises transmitting, by the first driver, the I/O operation to the second driver of the virtual disk coordinator host.

12. The non-transitory computer readable medium of claim 9 , wherein the performing comprises updating the database to set the first host to be the virtual disk coordinator host of the virtual disk object.

13. The non-transitory computer readable medium of claim 9 , wherein the executing of the I/O operation by communicating with hosts that are the coordinators for the virtual component objects of the virtual disk object comprises:
identifying, by the virtual disk coordinator host, virtual component objects of the virtual disk that are subject to the I/O operation, wherein a virtual component object encompasses a data partition of the virtual disk;
determining a component object coordinator host for each of the virtual component objects, wherein the component object coordinator host controls access to an associated virtual component object;
transmitting, by the virtual disk coordinator host, the I/O operation to each of the component object coordinator hosts; and
performing, by each of the component object coordinator hosts, the I/O operation in a local storage resource backing the virtual component object.

14. The non-transitory computer readable medium of claim 13 , wherein each host computer in the cluster contributes at least a portion of a corresponding local storage disk to form a pool of shared resources accessible as the virtual disk.

15. The non-transitory computer readable medium of claim 14 , wherein
at least one section of the virtual disk is backed by a local storage disk housed in the first host, 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 non-transitory computer readable medium of claim 15 ,
wherein the virtu

Assignees

• Vmware Inc

Inventors

• Earl William
• Karamanolis Christos

Classifications

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• H04L47/83

  based on usage prediction · CPC title

• G06F3/0653

  Monitoring storage devices or systems · CPC title

• G06F3/0605

  by facilitating the interaction with a user or administrator · CPC title

• G06F9/4411

  Configuring for operating with peripheral devices; Loading of device drivers · CPC title