Software overload control method

• US 5,440,741 A
• Filed: 09/20/1993
• Issued: 08/08/1995
• Est. Priority Date: 09/20/1993
• Status: Expired due to Term

• Current Assignee: Motorola Mobility LLC (Lenovo Group Ltd.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Morales, Javier, Hluchyj, Michael G., Humblet, Pierre A., Lee, Whay C.
• Primary Examiner(s): Heckler, Thomas M.
• Application Number:
  US08/123,943
• Time in Patent Office: 687 Days
• Field of Search: 395/200, 395/650, 340/825.5, 370/60.1, 370/85.4, 364/DIG. 1 MS File
• US Class Current: 718/103
• CPC Class Codes: G06F 11/004 Error avoidance G06F11/07 and subgroups take precedence
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 2209/5022 Workload threshold
• CPC Class Codes: G06F 9/505 considering the load

Abstract

A mechanism and method control software overloads in a communication network system to preserve the throughput capacity of the system in a simple and efficient manner. System load is regulated by accepting, deferring or rejecting work that the system is requested to perform. The criteria used to accept, defer or reject work is based on the relevance of the request and availability of system resources.

Claims

1. A software overload control mechanism for maintaining the throughput of a communication system, the communication system composed of a plurality of inter-linked nodes, and each node having a plurality of inter-linked communication processors, the communication system having system resources, a load and a capacity, the communication system being susceptible to an overload condition where the load exceeds the capacity of the communication system, by regulating the amount of work accepted by the communication system, the mechanism comprised of:

a rule-based decision maker for receiving a request from the communication system for work to be performed and determining whether the request should be;

rejected;

deferred;

or

executed;

an inventory monitor, for monitoring local and global system resource consumption coupled to the rule-based decision maker and

an overload scheduler coupled to the decision maker, for prioritizing and scheduling any request that is deferred.

2. The mechanism of claim 1 further comprising a resource manager for allocating and deallocating system resources based upon the determinations by the rule-based decision maker of the work to be performed.

3. The mechanism of claim 2 where the rule-based decision maker includes means for distinguishing the type of work to be performed by the system.

4. The mechanism of claim 3 where the rule-based decision maker is coupled to the inventory monitor by way of a probe for determining the overload condition of the system.

5. The mechanism of claim 4 where the probe is a relevance based request issued by the rule-based decision maker.

6. A communication system comprised of:

a plurality of inter-linked nodes, and each node having a plurality of inter-linked communication processors,

a plurality of message buffers coupled to the communication processors;

a rule-based decision maker for receiving a request from the communication system for the communication system to perform work and determining whether the request should be rejected, deferred;

or executed;

an inventory monitor, coupled to the rule-based decision maker, for monitoring local and global system utilization of message buffers; and

an overload scheduler coupled to the decision maker, for prioritizing and scheduling any request that is deferred.

7. The communication system of claim 6 where the rule-based decision maker includes distinguishing means for distinguishing the type of work to be performed by the system.

8. The communication system of claim 7 where the rule-based decision maker is coupled to the inventory monitor by way of a probe for determining the overload condition of the system.

9. The communication system of claim 8 where the probe is a relevance based request issued by the rule-based decision maker.

10. A method for maintaining the throughput of a communication system, the communication system composed of a plurality of inter-linked nodes, and each node has a plurality of inter-linked communication processors, the communication system being subject to overload, by regulating the amount of work accepted by the communication system comprising the steps of:

determining whether a request for work to be done by the communication system was made;

assessing the overload level of the communication system;

rejecting the request if the overload level of the system is above a first pre-determined level; and

accepting the request and executing the work if the overload level of the system is below a second pre-determined level.

11. The method of claim 10 where there are a plurality of different types of work, and where there are a plurality of different first and second pre-determined levels corresponding to each type of work.

12. The method of claim 11 where at least one type of work is to be performed regardless of overload level.

13. The method of claim 12 including the step of, if the work is not rejected, allocating system resources to perform the work immediately.

14. The method of claim 13 where the step of assessing the overload level of the system includes the step of assessing the overload level of at least one communication processor at the node.

15. The method of claim 13 including the step of assessing the overload level of all the communication processors at the node.

16. The method of claim 13 where the step of assessing the overload level of the system by assessing the overload level of all the communication processors at the node.

17. The method of claim 12 including the step of, if the work is not rejected, scheduling the work for execution by the system.