Identifying candidate workloads for migration

Patent metadata

Field	Value
Publication number	US-10445134-B2
Application number	US-201414295232-A
Country	US
Kind code	B2
Filing date	Jun 3, 2014
Priority date	Jun 3, 2014
Publication date	Oct 15, 2019
Grant date	Oct 15, 2019

Abstract

Techniques for identifying a candidate workload which may be a suitable candidate for migration from a first location to a second location are described herein. A set of suitability measurements associated with a computer system resource operating in the first location is received, the set of suitability measurements including, for example, resource usage values for one or more resources associated with the workload. Based at least in part on one or more statistical calculations on the set of suitability measurements exceeding one or more thresholds, recommendations are made about whether to migrate the workload from the first location to the second location.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
for each candidate workload of a plurality of candidate workloads:
receiving, from a client device, a resource usage profile associated with the candidate workload, the resource usage profile specifying a set of resource usage values associated with operating the candidate workload in a first operating environment, the set of resource usage values specifying one or more load measurements associated with one or more computer system resources, the first operating environment including computer hardware that is outside a control of a computer resource service provider;
determining one or more scaled resource usage metrics based at least in part on a coefficient of variation, the coefficient of variation calculated by dividing a standard deviation of a subset of the set of resource usage values in the resource usage profile by a mean of the subset of the set of resource usage values in the resource usage profile;
determining a migration recommendation for migrating the candidate workload from the first operating environment to a second operating environment based at least in part on:
whether the coefficient of variation of the one or more scaled resource usage metrics exceeds a threshold; and
a projected return on investment that is based at least in part on a comparison between:
a first operation cost of operating the candidate workload in the first operating environment; and
a second operation cost, the second operation cost being an estimated cost of operating the candidate workload in the second operating environment, wherein the second operation cost includes a migrated resource cost for migrating the candidate workload to the second operating environment, the second operating environment including virtualized computer instances that are provided as a service by the computing resource service provider; and

a determination of suitability of migration of the candidate workload based at least in part on a result of the projected return on investment attaining a value relative to a second threshold; and

providing, to be dynamically displayed in a graphical user interface of the client device, a representation of the candidate workload and the migration recommendation, the graphical user interface including functionality to initiate the migration of the candidate workload.

2. The computer-implemented method of claim 1 , wherein the one or more load measurements include at least one of: central processing unit usage, memory usage, disk input/output usage, or network usage.

3. The computer-implemented method of claim 1 , wherein the migration recommendation is based at least in part on a difference between the first operation cost and the second operation cost.

4. The computer-implemented method of claim 3 , wherein the second operation cost is based at least in part on costs associated with instantiating computer system resources in the second operating environment.

5. The computer-implemented method of claim 1 , wherein the first operation cost is based at least in part on at least one of:
a cost automatically generated by the computing resource service provider, or
a cost provided by an entity that manages the first operating environment.

6. The computer-implemented method of claim 1 , further comprising initiating migration of the candidate workload as a result of receiving a selection via the graphical user interface that initiates the migration of the candidate workload.

7. The computer-implemented method of claim 1 , wherein providing the representation includes providing a visualization that indicates that the migration is not recommended.

8. The computer-implemented method of claim 1 , wherein as a result of the projected return on investment exceeding a second threshold, suitability is determined to be high.

9. The computer-implemented method of claim 8 , wherein providing the representation includes providing a visualization that indicates that the suitability of the migration is high.

10. A system, comprising:
at least one computing device that implements one or more services of a computing resource service provider, wherein, for each candidate workload of a plurality of candidate workloads, the one or more services:
receive, from a client device, a resource usage profile associated with the candidate workload, the resource usage profile specifying a set of resource usage values associated with operating the candidate workload in a first operating environment, the set of resource usage values specifying one or more load measurements associated with one or more computer system resources, the first operating environment including computer hardware that is outside a control of a computer resource service provider;
determine one or more scaled resource usage metrics based at least in part on a coefficient of variation, the coefficient of variation calculated by dividing a standard deviation of a subset of the set of resource usage values in the resource usage profile by a mean of the subset of the set of resource usage values in the resource usage profile;
determine a migration recommendation for migration of the candidate workload from the first operating environment to a second operating environment based at least in part on:
whether the coefficient of variation of the one or more scaled resource usage metrics exceeds a threshold; and
a projected return on investment that is based at least in part on a comparison between:
a first operation cost of operating the candidate workload in the first operating environment; and
a second operation cost, the second operation cost being an estimated cost of operating the candidate workload in the second operating environment, wherein the second operation cost includes a migrated resource cost for the migration to the second operating environment, the second operating environment including virtualized computer instances that are provided as a service by the computing resource service provider; and

a determination of suitability of the migration based at least in part on a result of the projected return on investment attaining a value relative to a second threshold; and

provide, to be dynamically displayed in a graphical user interface of the client device, a representation of the candidate workload and the migration recommendation, the graphical user interface including functionality to initiate the migration of the candidate workload.

11. The system of claim 10 , wherein the one or more load measurements specify a current load value as a percentage of a maximum load value for the one or more computer system resources associated with the set of resource usage values.

12. The system of claim 10 , wherein the one or more load measurements are further based at least in part on calculating an absolute value of a difference between the set of resource usage values and an optimal usage value for the one or more computer system resources associated with the set of resource usage values.

13. The system of claim 10 , wherein measurement is based at least in part on a difference between a first set of cost calculations associated with operating the candidate workload in the first operating environment and a second set of cost calculations associated with operating the candidate workload in the second operating environment.

14. The system of claim 13 , wherein the first set of cost calculations associated with operating the candidate workload in the first operating environment are based at least in part on a costs of ownership associated with computer system equipment in the first operating environment.

Assignees

• Amazon Tech Inc

Inventors

• Borthakur Apolak

Classifications

• G06F9/4856 Primary

  resumption being on a different machine, e.g. task migration, virtual machine migration (G06F9/5088 takes precedence) · CPC title

• G06F9/5088 Primary

  involving task migration · CPC title