Multi-stage IOPS allocation

Patent metadata

Field	Value
Publication number	US-11494241-B2
Application number	US-202117171195-A
Country	US
Kind code	B2
Filing date	Feb 9, 2021
Priority date	Oct 31, 2018
Publication date	Nov 8, 2022
Grant date	Nov 8, 2022

Abstract

Systems and methods for policy-based apportionment of input/output operations (IOPS) in computing systems. Embodiments access a policy that specifies IOPS limits. Two or more virtual machines that are associated with the policy and two or more nodes that host those virtual machines are identified. In a first allocation stage, an inter-node policy manager prescribes an initial IOPS limit to the two or more nodes. The allocation amounts sent to the nodes depend at least in part on performance capabilities of respective nodes. In a second allocation stage, for each node that had received a limit amount, that amount is apportioned to the sets of virtual machines that execute on respective host nodes. Each node of the two or more nodes invokes its own node-local IOPS monitoring. Each node reports IOPS usage data to the inter-node policy manager, which in turn adjusts the node-level IOPS apportionments based on the node-level usage.

First claim

What is claimed is:

1. A non-transitory computer-readable medium having stored thereon a sequence of instructions which, when executed by a processor, causes a set of acts comprising:
identifying multiple nodes associated with a policy, the policy specifying an input/output operations per second (IOPS) limit;
allocating a portion of IOPS of the IOPS limit to each of the multiple nodes, wherein a respective portion of the IOPS allocated to a respective node of the multiple nodes is to be shared across at least one virtualized entity of the respective node, wherein an IOPS usage monitoring function at two or more nodes of the multiple nodes measures node-specific IOPS usage;
receiving node-specific IOPS usage data for the two or more nodes comprising measured IOPS; and
reallocating at least a first subset of a first portion of IOPS allocated to a first node of the two or more nodes based on at least the measured IOPS for the first node, wherein the measured IOPS for the first node is different from the portion of IOPS previously allocated to the first node and the first subset is reallocated to a second node.

2. The computer-readable medium of claim 1 , wherein the two or more nodes comprises at least the first and second nodes, and reallocation of the first subset to the second node increases a second portion of IOPS allocated to the second node.

3. The computer-readable medium of claim 1 , wherein the measured IOPS is used to generate a predicted IOPS rate.

4. The computer-readable medium of claim 3 , wherein the predicted IOPS rate is derived from measured IOPS rates over multiple observation periods.

5. The computer-readable medium of claim 1 , wherein the set of acts further comprise determining whether a change is to be made to a portion of IOPS to be allocated to the node in response to a change in the policy.

6. The computer-readable medium of claim 5 , wherein the change to be made to the portion of IOPS to be allocated to the node is determined based on at least the measured IOPS and the change in the policy.

7. The computer-readable medium of claim 1 , wherein the set of acts further comprise throttling one or more I/O operations from respective virtualized entities of a plurality of virtualized entities on respective nodes of the multiple nodes, the throttling being based at least in part on at least the IOPS allocated to a respective node.

8. The computer-readable medium of claim 1 , wherein the IOPS limit comprises at least one of, an aggregate IOPS limit, or a provisioned IOPS limit.

9. A method comprising:
identifying multiple nodes associated with a policy, the policy specifying an input/output operations per second (IOPS) limit;
allocating a portion of IOPS of the IOPS limit to each of the multiple nodes, wherein a respective portion of the IOPS allocated to a respective node of the multiple nodes is to be shared across at least one virtualized entity of the respective node, wherein an IOPS usage monitoring function at two or more nodes of the multiple nodes measures node-specific IOPS usage;
receiving node-specific IOPS usage data for the two or more nodes comprising measured IOPS; and
reallocating at least a first subset of a first portion of IOPS allocated to a first node of the two or more nodes based on at least the measured IOPS for the first node, wherein the measured IOPS for the first node is different from the portion of IOPS previously allocated to the first node and the first subset is reallocated to a second node.

10. The method of claim 9 , wherein the two or more nodes comprises at least the first and second nodes, and reallocation of the first subset to the second node increases a second portion of IOPS allocated to the second node.

11. The method of claim 9 , wherein the measured IOPS is used to generate a predicted IOPS rate.

12. The method of claim 11 , wherein the predicted IOPS rate is derived from measured IOPS rates over multiple observation periods.

13. The method of claim 9 , further comprising determining whether a change is to be made to a portion of IOPS to be allocated to the node in response to a change in the policy.

14. The method of claim 13 , wherein the change to be made to the portion of IOPS to be allocated to the node is determined based on at least the measured IOPS and the change in the policy.

15. The method of claim 9 , further comprising throttling one or more I/O operations from respective virtualized entities of a plurality of virtualized entities on respective nodes of the multiple nodes, the throttling being based at least in part on at least the IOPS allocated to a respective node.

16. The method of claim 9 , wherein the IOPS limit comprises at least one of, an aggregate IOPS limit, or a provisioned IOPS limit.

17. A system for multi-stage input/output operations (IOPS) allocations, the system comprising:
a storage medium having stored thereon a set of instructions; and
a processor that executes the set of instructions to cause a set of acts, the set of acts comprising:
identifying multiple nodes associated with a policy, the policy specifying an input/output operations per second (IOPS) limit;
allocating a portion of IOPS of the IOPS limit to each of the multiple nodes, wherein a respective portion of the IOPS allocated to a respective node of the multiple nodes is to be shared across at least one virtualized entity of the respective node, wherein an IOPS usage monitoring function at two or more nodes of the multiple nodes measures node-specific IOPS usage;
receiving node-specific IOPS usage data for the two or more nodes comprising measured IOPS; and
reallocating at least a first subset of a first portion of IOPS allocated to a first node of the two or more nodes based on at least the measured IOPS for the first node, wherein the measured IOPS for the first node is different from the portion of IOPS previously allocated to the first node and the first subset is reallocated to a second node.

18. The system of claim 17 , wherein the two or more nodes comprises at least the first and second nodes, and reallocation of the first subset to the second node increases a second portion of IOPS allocated to the second node.

19. The system of claim 17 , wherein the measured IOPS is used to generate a predicted IOPS rate.

20. The system of claim 19 , wherein the predicted IOPS rate is derived from measured IOPS rates over multiple observation periods.

21. The system of claim 17 , wherein the set of acts further comprise determining whether a change is to be made to a portion of IOPS to be allocated to the node in response to a change in the policy.

22. The system of claim 21 , wherein the change to be made to the portion of IOPS to be allocated to the node is determined based on at least the measured IOPS and the change in the policy.

23. The system of claim 17 , wherein the set of acts further comprise throttling one or more I/O operations from respective virtualized entities of a plurality of virtualized entities on respective nodes of the multiple nodes, the throttling being based at least in part on at least the IOPS allocated to a respective node.

24. The system of claim 17 , wherein the IOPS limit comprises at least one of, an aggregate IOPS limit, or a provisioned IOPS limit.

Assignees

• Nutanix Inc

Inventors

• Jain Kshitiz
• Kajaria Prateek

Classifications

• G06F2009/45595

  Network integration; Enabling network access in virtual machine instances · CPC title

• G06F9/45558

  Hypervisor-specific management and integration aspects · CPC title

• G06F2009/45579

  I/O management, e.g. providing access to device drivers or storage · CPC title

• G06F9/5077 Primary

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F2009/45583

  Memory management, e.g. access or allocation · CPC title