System and method for risk-aware server load balancing

Patent metadata

Field	Value
Publication number	US-10476950-B2
Application number	US-201615288481-A
Country	US
Kind code	B2
Filing date	Oct 7, 2016
Priority date	Oct 9, 2015
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

Systems and methods are provided for managing server loads that accounts for various measures of risk associated with different workloads assigned to servers. The systems and methods may include a memory storing instructions for server load management operations, and a processor configured to execute the stored instructions. The processor may receive a workload, determine a value associated with the workload indicating a predetermined importance of the workload, receive information for a plurality of active servers in a server cluster associated with the processor, determine risk levels associated with the active servers based on the received information, and assign the received workload to one of the active servers based on the determined value and the determined risk levels.

First claim

What is claimed is:

1. A load balancer system for managing server loads for an organization, the load balancer system comprising:
a memory storing instructions for server load management operations; and
a processor configured to execute the stored instructions, to:
receive a workload from a client device;
determine a cumulative workload value associated with the workload, the cumulative workload value indicating an importance of the workload, relative to other workloads, to success of the organization based on at least one of:
a monetary value associated with the workload, a risk of degradation in prestige of the organization associated with the workload, or a risk of loss of customers of the organization associated with the workload;

receive information for a plurality of active servers in a server cluster associated with the processor, the information including a failure rate for each of the active servers and a real-time resource load status for each of the active servers;
receive, from a database associated with the load balancer system, a policy comprising a cumulative workload value limit for each of the active servers;
determine risk levels associated with the active servers based on the received information; and
assign the received workload to one of the active servers based on the cumulative workload value, the received policy, and the determined risk levels.

2. The load balancer system of claim 1 , wherein:
the processor is further configured to determine a resource load associated with one or more workloads currently assigned to each of the active servers, and
the received workload is assigned to one of the active servers based at least in part on the determined resource load.

3. The load balancer system of claim 1 , wherein the processor is further configured to monitor a real-time risk level for each of the active servers.

4. The load balancer system of claim 1 , wherein:
the processor generates a candidate server list having at least one of the active servers, the candidate server list being determined based on the determined risk levels, and
the received workload is assigned to a server from the candidate server list.

5. The load balancer system of claim 4 , wherein the active servers are ranked based on the determined risk levels, and the candidate server list comprises a predetermined number of ranked active servers having a lowest determined risk level.

6. The load balancer system of claim 4 , wherein the processor evaluates the received workload by predicting a hypothetical effect of the received workload on each of the servers in the candidate server list.

7. The load balancer system of claim 6 , wherein the processor assigns the received workload based at least in part on the evaluation.

8. The load balancer system of claim 1 , wherein the policy further comprises at least one rule for migrating workloads between servers.

9. A method for managing server loads for an organization, comprising:
receiving a workload from a client device;
determining a cumulative workload value associated with the workload, the cumulative workload value indicating an importance of the workload, relative to other workloads, to success of the organization based on at least one of:
a monetary value associated with the workload, a risk of degradation in prestige of the organization associated with the workload, or a risk of loss of customers of the organization associated with the workload;

receiving information for a plurality of active servers in a server cluster associated with the processor, the information including a failure rate for each of the active servers and a real-time resource load status for each of the active servers;
receiving, from a database associated with the load balancer system, a policy comprising a cumulative workload value limit for each of the active servers;
determining risk levels associated with the active servers based on the received information; and
assigning the received workload to one of the active servers based on the cumulative workload value, the received policy, and the determined risk levels.

10. The method of claim 9 , further comprising:
determining a resource load associated with one or more workloads currently assigned to each of the active servers,
wherein the received workload is assigned to one of the active servers based at least in part on the determined resource load.

11. The method of claim 9 , further comprising maintaining, in a database, a real-time risk level for each of the active servers.

12. The method of claim 9 , further comprising:
generating a candidate server list having at least one of the active servers, the candidate server list being determined based on the determined risk levels,
wherein the received workload is assigned to a server from the candidate server list.

13. The method of claim 12 , wherein the active servers are ranked based on the determined risk levels, and the candidate server list comprises a predetermined number of ranked active servers having a lowest determined risk level.

14. The method of claim 12 , wherein the processor evaluates the received workload by predicting a hypothetical effect of the received workload on each of the servers in the candidate server list.

15. The method of claim 14 , wherein the processor assigns the received workload based at least in part on the evaluation.

16. The method of claim 9 , wherein the policy further comprises at least one rule for migrating workloads between servers.

17. A non-transitory computer readable medium having stored instructions, which when executed, cause at least one processor to perform operations for managing server loads for an organization, comprising:
receiving a workload from a client device;
determining a cumulative workload importance value associated with the workload, the cumulative workload value indicating an importance of the workload, relative to other workloads, to success of the organization based on at least one of:
a monetary value associated with the workload, a risk of degradation in prestige of the organization associated with the workload, or a risk of loss of customers of the organization associated with the workload;

receiving information for a plurality of active servers in a server cluster associated with the processor, the information including a failure rate for each of the active servers and a real-time resource load status for each of the active servers;
receiving, from a database, a policy comprising a cumulative workload value limit for each of the active servers;
determining risk levels associated with the active servers based on the received information; and
assigning the received workload to one of the active servers based on the cumulative workload value, the received policy, and the determined risk levels.

Assignees

• Capital One Services Llc

Inventors

• Tao Tao
• Bardwaj Santosh
• Yoo Il Sun
• Tang Yihui
• GERSTLE Jeremy

Classifications

• H04L67/1029 Primary

  using data related to the state of servers by a load balancer · CPC title