Vertical scaling of computing instances

Patent metadata

Field	Value
Publication number	US-10355934-B2
Application number	US-201414559823-A
Country	US
Kind code	B2
Filing date	Dec 3, 2014
Priority date	Dec 3, 2014
Publication date	Jul 16, 2019
Grant date	Jul 16, 2019

Abstract

Technology for scaling computing instances is provided. In one example, a method may include identifying vertical scaling rules defining a criterion for vertically scaling a computing instance from a first instance type to a second instance type. The method may further include monitoring usage of the computing instance and vertically scaling the computing instance from the first instance type to the second instance type based on the usage as compared to the vertical scaling rules.

First claim

The invention claimed is:

1. A computing device that is configured to scale a group of computing instances, comprising:
a processor;
a memory in electronic communication with the processor;
instructions stored in the memory, the instructions being executable by the processor to:
identify vertical scaling rules defining criteria for circumstances to vertically scale at least two computing instances managed by a service provider environment from a first computing instance type to a second computing instance type, wherein the first computing instance type and the second computing instance type are pre-defined computing instance types specifying a different pre-defined configuration of computing resources offered by the service provider environment as a computing instance;
monitor a load on the at least two computing instances running on one or more physical hosts of the service provider environment;
vertically scale the at least two computing instances from the first computing instance type to the second computing instance type when the load on the at least two computing instances reaches a load indicated by at least one of the vertical scaling rules associated with the at least two computing instances and defining how a first pre-defined computing instance type is changed to a second pre-defined computing instance type; and
horizontally scale the at least two computing instances to change a number of computing instances based, at least in part, on the second computing instance type.

2. The computing device of claim 1 , further configured to evaluate the number of computing instances and the first instance type together to identify the second instance type from among a plurality of different instance types for vertically scaling the at least two computing instances up or down to respectively larger or smaller computing instance types based on the load.

3. The computing device of claim 1 , further configured to replace an entirety of a group of computing instances of the first instance type with a group of computing instances of the second instance type.

4. The computing device of claim 3 , further configured to perform a rolling replacement of a plurality of computing instances of the first instance type, wherein different computing instances are replaced at different times.

5. The computing device of claim 1 , wherein vertically scaling the at least two computing instances comprises replacing less than an entirety of the number of computing instances of the first instance type with computing instances of the second instance type.

6. A computer-implemented method, comprising:
identifying vertical scaling rules defining a criterion for vertically scaling a plurality of computing instances managed by a service provider environment from a first instance type to a second instance type, wherein the first instance type and the second instance type are pre-defined computing instance types specifying a different pre-defined configuration of computing resources offered by the service provider environment as a computing instance;
monitoring usage of the plurality of computing instances running on one or more physical hosts of the service provider environment, using a processor; and
vertically scaling the plurality of computing instances from the first instance type to the second instance type based, based at least in part, on usage data as compared to at least one of the vertical scaling rules associated with the plurality of computing instances defining how a first pre-defined computing instance type is changed to a second pre-defined computing instance type.

7. The method of claim 6 , further comprising selecting the second instance type from among a plurality of different instance types based on whether vertical scaling includes downgrading the plurality of computing instances or upgrading the plurality of computing instances.

8. The method of claim 6 , further comprising evaluating a cost of maintaining the plurality of computing instances in combination with the usage of the plurality of computing instances in view of the vertical scaling rules to identify the second instance type.

9. The method of claim 6 , further comprising vertically scaling the plurality of computing instances upward to the second instance type, the second instance type having increased computing performance characteristics as compared with the first instance type.

10. The method of claim 6 , further comprising vertically scaling the plurality of computing instances downward to the second instance type, the second instance type having decreased computing performance characteristics as compared with the first instance type.

11. The method of claim 6 , wherein the vertical scaling rules are configured to implement vertical scaling to minimize cost while maintaining a defined performance level of multiple of the plurality of computing instances as a group.

12. The method of claim 6 , wherein the vertical scaling rules comprise a scaling alarm to notify an administrator when the usage of the plurality of computing instances exceeds a threshold and a scaling trigger to initiate vertically scaling the plurality of computing instances.

13. The method of claim 6 , further comprising horizontally scaling the plurality of computing instances to a smaller number of computing instances when vertically scaling the plurality of computing instances to a larger type of computing instance or horizontally scaling the plurality of computing instances to a larger number of computing instances when vertically scaling the plurality of computing instances to a smaller type of computing instance.

14. The method of claim 6 , wherein the instance type is identified by at least one type indicator selected from the group consisting of: instance type name, memory, compute units, storage, architecture, I/O performance and cost.

15. The method of claim 6 , wherein the vertical scaling rules are based, at least in part, on a relationship between one or more variables associated with the first instance type and the second instance type.

16. The method of claim 15 , wherein the relationship is a ratio of a number of computing instances of the first instance type compared to a number of computing instances of the second instance type.

17. The method of claim 6 , further comprising maintaining operation of a defined minimum number of the plurality of computing instances during vertical scaling.

18. The method of claim 6 , wherein the vertical scaling further comprises horizontal scaling to increase or decrease a number of the plurality of computing instances.

19. A non-transitory computer-readable medium comprising computer-executable instructions which, when executed by a processor, implement a system, comprising:
a rules data store configured to store vertical scaling rules defining a second instance type to which to vertically scale computing instances managed by a service provider environment from a first instance type, wherein the first instance type and the second instance type are pre-defined computing instance types associated with the service provider environment specifying a different pre-defined configuration of computing resources offered by the service provider environment as a computing instance;
a monitoring agent configured to monitor a load on the computing instances running on one or more physical hosts of the service provider environment; and
a vertical scaling module configured to vertically scale the computing instances from the first instance type to the second instance type when the load

Assignees

• Amazon Tech Inc

Inventors

• Pinski Nikita
• Pai Derek Solomon
• Khalid Ahmed Usman
• Stone Eric Samuel
• Bukovec Mai-Lan Tomsen

Classifications

• H04L41/0897

  by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities · CPC title

• H04L41/0896 Primary

  Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities (flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76) · CPC title

• G06F9/5061 Primary

  Partitioning or combining of resources · CPC title

• H04L43/0888

  Throughput · CPC title