Method and system for scheduling input/output resources of a virtual machine

1. A computer-implemented method for scheduling I/O resources of a virtual machine, the method comprising:

configuring a scheduling strategy based on a priority for each of a plurality of I/O queues, wherein a respective I/O queue corresponds to a virtual disk of the virtual machine, and wherein a respective virtual disk is stored as a file on a physical machine hosting the virtual machine;

maintaining the plurality of I/O queues based on the scheduling strategy and a portion of shared I/O resources allocated to groups corresponding to the plurality of I/O queues,

wherein a respective group includes a plurality of service trees, wherein a respective service tree includes a first number of the I/O queues, and wherein the first number of the I/O queues comprises a unique part of an entirety of the plurality of I/O queues;

assigning an I/O request to a first I/O queue based on the scheduling strategy and the allocated portion of the shared I/O resources; and

allocating, to the I/O request, a first amount of the portion of the shared I/O resources, wherein the first amount represents a quota for the virtual disk corresponding to the first I/O queue.

2. The method of claim 1 , wherein the groups, the service trees, and the I/O queues are arranged in a tree-like structure with the I/O queues arranged as leaves and the groups arranged as roots,

wherein a respective service tree includes a unique set of the plurality of I/O queues, and

wherein a respective group, a respective service tree, and a respective unique set of the plurality of I/O queues are determined based on a completely fair queuing scheduling algorithm.

3. The method of claim 1 , further comprising:

receiving the I/O request; and

obtaining one or more features from the I/O request.

4. The method of claim 3 , wherein a feature includes one or more of:

a first value which is set by a user of the computer system;

a second value which is set by the computer system; and

a predetermined value, and

wherein the method further comprises:

in response to determining that the feature includes a first value which is set by the user, marking the I/O request.

5. The method of claim 3 , wherein a feature includes a process identifier which indicates whether a source of the I/O request is a virtual machine or a physical machine, and wherein the method further comprises:

in response to determining that the feature includes a process identifier which indicates that the source of the I/O request is a virtual machine, identifying the virtual disk stored as the file identified by a file identifier contained in the I/O request.

6. The method of claim 1 , wherein assigning the I/O request to the first I/O queue is further based on a file identifier contained in the I/O request and scheduling parameters for the virtual disk,

wherein the file identifier uniquely identifies the file storing the virtual disk corresponding to the first I/O queue, and

wherein the scheduling parameters indicate a portion of the shared I/O resources allocated to the virtual disk.

7. The method of claim 1 , further comprising:

setting a first threshold which indicates a value associated with triggering a re-allocation of the shared I/O resources; and

in response to detecting that a virtual disk is idle and has an allocated portion of the shared I/O resources greater than the first threshold, re-allocating the shared I/O resources for virtual disks of the virtual machine.

8. The method of claim 1 , further comprising:

determining a target virtual disk of the virtual machine based on a file identifier contained in the I/O request; and

determining a second I/O queue which is mapped to the target virtual disk, wherein the second I/O queue is the first I/O queue to which the I/O request is assigned.

9. The method of claim 1 , wherein assigning the I/O request to the first I/O queue causes virtual disks of the virtual machine to remain isolated from each other.

10. The method of claim 1 , wherein the I/O request is received from a user of one or more of:

a first physical computing device;

a first virtual machine associated with the first physical computing device;

a second physical computing device which interacts with the first physical computing device or the first virtual machine; and

a second virtual machine which interacts with the first physical computing device or the first virtual machine.

11. A computer system for scheduling I/O resources of a virtual machine, the system comprising:

a processor; and

a memory coupled to the processor and storing instructions, which when executed by the processor cause the processor to perform a method, the method comprising:

configuring a scheduling strategy based on a priority for each of a plurality of I/O queues, wherein a respective I/O queue corresponds to a virtual disk of the virtual machine, and wherein a respective virtual disk is stored as a file on a physical machine hosting the virtual machine;

maintaining the plurality of I/O queues based on the scheduling strategy and a portion of shared I/O resources allocated to groups corresponding to the plurality of I/O queues,

wherein a respective group includes a plurality of service trees, wherein a respective service tree includes a first number of the I/O queues, and wherein the first number of the I/O queues comprises a unique part of an entirety of the plurality of I/O queues;

assigning an I/O request to a first I/O queue based on the scheduling strategy and the allocated portion of the shared I/O resources; and

allocating, to the I/O request, a first amount of the portion of the shared I/O resources, wherein the first amount represents a quota for the virtual disk corresponding to the first I/O queue.

12. The computer system of claim 11 , wherein the groups, the service trees, and the I/O queues are arranged in a tree-like structure with the I/O queues arranged as leaves and the groups arranged as roots,

wherein a respective service tree includes a unique set of the plurality of I/O queues, and

wherein a respective group, a respective service tree, and a respective unique set of the plurality of I/O queues are determined based on a completely fair queuing scheduling algorithm.

13. The computer system of claim 11 , wherein the method further comprises:

receiving the I/O request; and

obtaining one or more features from the I/O request.

14. The computer system of claim 13 , wherein a feature includes one or more of:

a first value which is set by a user of the computer system;

a second value which is set by the computer system; and

a predetermined value, and

wherein the method further comprises:

in response to determining that the feature includes a first value which is set by the user, marking the I/O request.

15. The computer system of claim 13 , wherein a feature includes a process identifier which indicates whether a source of the I/O request is a virtual machine or a physical machine, and wherein the method further comprises:

in response to determining that the feature includes a process identifier which indicates that the source of the I/O request is a virtual machine, identifying the virtual disk stored as the file identified by a file identifier contained in the I/O request.

16. The computer system of claim 11 , wherein assigning the I/O request to the first I/O queue is further based on a file identifier contained in the I/O request and scheduling parameters for the virtual disk,

wherein the file identifier uniquely identifies the file storing the virtual disk corresponding to the first I/O queue, and

wherein the scheduling parameters indicate a portion of the shared I/O resources allocated to the virtual disk.

17. The computer system of claim 11 , wherein the method further comprises:

setting a first threshold which indicates a value associated with triggering a re-allocation of the shared I/O resources; and

in response to detecting that a virtual disk is idle and has an allocated portion of the shared I/O resources greater than the first threshold, re-allocating the shared I/O resources for virtual disks of the virtual machine.

18. The computer system of claim 11 , wherein the method further comprises:

determining a target virtual disk of the virtual machine based on a file identifier contained in the I/O request; and

determining a second I/O queue which is mapped to the target virtual disk, wherein the second I/O queue is the first I/O queue to which the I/O request is assigned.

19. The computer system of claim 11 , wherein assigning the I/O request to the first I/O queue causes virtual disks of the virtual machine to remain isolated from each other.

20. The computer system of claim 11 , wherein the I/O request is received from a user of one or more of:

a first physical computing device;

a first virtual machine associated with the first physical computing device;

a second physical computing device which interacts with the first physical computing device or the first virtual machine; and

a second virtual machine which interacts with the first physical computing device or the first virtual machine.