Usage-based server load balancing

Patent metadata

Field	Value
Publication number	US-11336718-B2
Application number	US-202016947008-A
Country	US
Kind code	B2
Filing date	Jul 15, 2020
Priority date	Feb 11, 2019
Publication date	May 17, 2022
Grant date	May 17, 2022

Abstract

A load balancer determines a first usage load for a first server group that is one of a plurality of server groups associated with a resource. The load balancer determines a usage total for a user group of a plurality of user groups assigned to make requests for the resource via the first server group. The load balancer determines an assignment of the user group to make requests for the resource via the first server group or a second server group of the plurality of server groups based on the usage total of the user group, the first usage load of the first server group, and a second usage load of the second server group. The load balancer routes requests for the resource by the user group to the first server group or the second server group based on the assignment.

First claim

What is claimed is:

1. A method, comprising:
determining, by a device, a first usage load for a first server group associated with a resource, and
wherein the first usage load is an aggregate usage total of the first server group over a time period by a plurality of user groups assigned to make requests for the resource via the first server group;

determining, by the device, a usage total for a user group of the plurality of user groups assigned to make requests for the resource via the first server group,
wherein the usage total identifies a usage of the first server group by the user group over the time period;

determining, by the device, that an assignment of the user group to make requests for the resource is to a second server group according to a difference of the first usage load and the usage total being within a range of a usage load threshold,
wherein the difference corresponds to when the first usage load is decreased by the usage total; and

selectively routing, by the device, a request for the resource by the user group to the first server group or the second server group according to the assignment.

2. The method of claim 1 , wherein the usage total is determined based on a weighting of data associated with at least one of:
a process duration of requests made by the user group over the time period; or
a quantity of requests made by the user group over the time period.

3. The method of claim 1 , wherein the usage total of the first server group by the user group over the time period is determined based on a weighting of one or more of:
a first total processing time for requests of the user group over the time period,
a first total number of requests by the user group over the time period,
a second total processing time for requests of the user group over an interval of the time period,
a second total number of requests by the user group over the interval of the time period,
a standard deviation in total processing time for requests of the user group over a plurality of intervals of the time period,
a standard deviation in total number of requests by the user group over the plurality of intervals of the time period,
an average total processing time for requests of the user group over the plurality of intervals of the time period, or
an average total number of requests by the user group over the plurality of intervals of the time period.

4. The method of claim 1 , further comprising:
determining respective usage totals for the plurality of user groups;
classifying the plurality of user groups into one of:
a light user group that satisfies a light user group threshold, or
a heavy user group that satisfied a heavy user group threshold; and

isolating heavy user groups from one another.

5. The method of claim 1 , wherein determining whether the assignment of the user group to make requests for the resource is via the second server group comprises:
determining the assignment of the user group to make requests for the resource via the second server group by employing an iterative process; and
the method further comprising:
recording a state of the iterative process.

6. The method of claim 1 , wherein the first server group is associated with a first plurality of server devices and the second server group is associated with a second plurality of server devices, and
wherein the first plurality of server devices and the second plurality of server devices do not have a server device in common.

7. The method of claim 1 , wherein the first server group is associated with a first plurality of server devices and the second server group is associated with a second plurality of server devices, and
wherein the first plurality of server devices and the second plurality of server devices have at least one server device in common.

8. A device, comprising:
one or more memories; and
one or more processors, communicatively coupled to the one or more memories, configured to:
determine a first usage load that is an aggregate usage total of a first server group over a time period by a plurality of user groups assigned to make requests for a resource via the first server group,
wherein the first server group is one of a plurality of server groups associated with the resource;

determine a usage total for a user group of the plurality of user groups assigned to make requests for the resource via the first server group,
wherein the usage total identifies a usage of the first server group by the user group over the time period;

determine that an assignment of the user group to make requests for the resource is to a second server group according to a difference of the first usage load and the usage total being within a range of a usage load threshold,
wherein the difference corresponds to when the first usage load is decreased by the usage total; and

selectively route a request for the resource by the user group to the first server group or the second server group according to the assignment.

9. The device of claim 8 , wherein the usage total is determined based on a weighting of data associated with at least one of:
a process duration of requests made by the user group over the time period; or
a quantity of requests made by the user group over the time period.

10. The device of claim 8 , wherein the first server group is associated with a first plurality of server devices and the second server group is associated with a second plurality of server devices,
wherein the first plurality of server devices and the second plurality of server devices have at least one server device in common.

11. The device of claim 8 , wherein the first server group is associated with a first plurality of server devices and the second server group is associated with a second plurality of server devices,
wherein the first plurality of server devices and the second plurality of server devices do not have a server device in common.

12. The device of claim 8 , wherein the usage of the first server group by the user group over the time period is based on determining a weighting of one or more of:
a first total processing time for requests of the user group over the time period,
a first total number of requests by the user group over the time period,
a second total processing time for requests of the user group over an interval of the time period,
a second total number of requests by the user group over the interval of the time period,
a standard deviation in total processing time for requests of the user group over a plurality of intervals of the time period,
a standard deviation in total number of requests by the user group over the plurality of intervals of the time period,
an average total processing time for requests of the user group over the plurality of intervals of the time period, or
an average total number of requests by the user group over the plurality of intervals of the time period.

13. The device of claim 8 , wherein the one or more processors are further to:
determine respective usage totals for the plurality of user groups;
classify the plurality of user groups into one of:
a light user group that satisfies a light user group threshold, or
a heavy user group that satisfied a heavy user group threshold; and

isolate heavy user groups from one another.

14. The device of claim 13 , wherein the one or more processors are further to:
determine assignments for the heavy user groups,
wherein a first heavy group, of the heavy user groups, is assigned to the first server group and a second heavy group, of the heavy user groups, is assigned to the second server group based on determ

Assignees

• Verizon Patent & Licensing Inc

Inventors

• James Matthew W
• Knaack Joseph B
• Holtfrerich David
• Lin Lei
• Lindee Jessica Marie

Classifications

• H04L67/1023

  based on a hash applied to IP addresses or costs · CPC title

• H04L67/101

  based on network conditions · CPC title

• H04L67/1044

  Group management mechanisms  (management of multicast group membership H04L12/185; reconfiguring of node membership in a computing system to eliminate errors G06F11/1425) · CPC title

• H04L67/1008 Primary

  based on parameters of servers, e.g. available memory or workload (monitoring of computer activity G06F11/30) · CPC title

• H04L67/1014 Primary

  based on the content of a request · CPC title