Predictive model-based measurement acquisition

• US 20020099821A1
• Filed: 03/29/2002
• Published: 07/25/2002
• Est. Priority Date: 03/13/1998
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Haus, Nikolaus, Hellerstein, Joseph L.
• Granted Patent: US 6,678,730 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: H04L 41/0233 Object-oriented techniques, for representation of network management data, e.g. common object request broker architecture [CORBA]
• CPC Class Codes: H04L 41/046 comprising network management agents or mobile agents therefor
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/022 by sampling
• CPC Class Codes: H04L 43/106 using time related information in packets, e.g. by adding timestamps
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1023 based on a hash applied to IP addresses or costs
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer

Abstract

A method, system and program storage device for measurement acquisition using predictive models which: (a) can improve scalability; (b) can accommodate devices that operate in a disconnected mode; and (c) enable integration of data from sources with different time granularities. Various features can be embodied in software and an object-oriented implementation is described. Different implementations are described, such as standalone predictive models implemented only on a manager (for example systems management/load balancing) or managed system (for example router management); or a parallel implementation with predictive models running on both the manager and managed (agent) systems (for example financial trading or system health monitoring). In a parallel model implementation, the agent constructs a predictive model that is conveyed to the manager system. The models are used in parallel, possibly with no communication for an extended time. The manager uses its model to provide tentative values of measurement variables to management applications. The agent uses its model to check its accuracy. If the model is found to be insufficiently accurate, an updated model is transmitted to the manager. The invention allows other measurement acquisition protocols to operate concurrently on the same measurement variables.

Claims

1. A predictive model-based method for measurement data acquisition in a data processing system, comprising the steps of:

dynamically constructing a predictive model for managing system resources, wherein the predictive model includes a model definition, estimates of model parameters, and input data; and

managing said system resources using predicted values for resource consumption generated by the predictive model.

2. A predictive model-based method for measurement data acquisition in a distributed system, comprising the steps of:

dynamically constructing a plurality of predictive models on a manager system, wherein each predictive model corresponds to a managed system and includes a model definition, estimates of model parameters, and model measurement variable input data; and

managing the managed systems based on predicted values from the predictive model.

3. The method of claim 2, further comprising the steps of:

said managing including load balancing work across the managed systems based on the predictive models for the managed systems.

4. A predictive model-based method for measurement data acquisition in a distributed system, comprising the steps of:

dynamically constructing a predictive model of pricing data, based on one or more non-static measurement variable values, on both a manager system and a disconnectable managed pricing-data system; and

the manager system processing requests for pricing data values based on the predictive model on the manager system.

5. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for predictive model-based method for measurement data acquisition in a data processing system, said method steps comprising:

dynamically constructing a predictive model for managing system resources, wherein the predictive model includes a model definition, estimates of model parameters, and input data; and

managing said system resources using predicted values for resource consumption generated by the predictive model.

6. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for predictive model-based method for measurement data acquisition in a distributed system, said method steps comprising:

dynamically constructing one or more predictive models on a manager system, wherein each predictive model corresponds to a managed system and includes a model definition, estimates of model parameters, and model measurement variable input data; and

managing the managed systems based on predicted values from the predictive model.

7. The method of claim 6, further comprising the steps of:

said managing including load balancing work across the managed systems based on the predictive models for the managed systems.

8. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for predictive model-based method for measurement data acquisition in a distributed system, said method steps comprising:

dynamically constructing a predictive model of pricing data, based on one or more non-static measurement variable values, on both a manager system and a disconnectable managed pricing-data system; and

the manager system processing requests for pricing data values based on the predictive model on the manager system.

9. A predictive model-based measurement acquisition system for creating and managing one or more predictive models that predict values of measurement variables, said system comprising:

means for dynamically constructing a predictive model for managing system resources, wherein the predictive model includes a model definition, estimates of model parameters, and input data; and

means for managing said system resources using predicted values for resource consumption generated by the predictive model.

10. The system of claim 9, wherein the system is a router which includes including a plurality of thresholds used to control CPU consumption by processes, further comprising:

the predictive model being adapted to predict future values of said CPU consumption; and

means for managing the thresholds based on predicted future values of said CPU consumption.

11. A predictive model-based measurement acquisition system for creating and managing one or more predictive models that predict values of measurement variables in distributed systems, said system comprising:

means for dynamically constructing one or more predictive models on a manager system, wherein each predictive model corresponds to a managed system and includes a model definition, estimates of model parameters, and model measurement variable input data; and

manager means for managing the managed systems based on predicted values from a predictive model on the manager system.

12. The method of claim 11, further comprising the steps of:

said manager means for managing including means for load balancing work across the managed systems based on the predictive model on the manager system.

13. A predictive model-based measurement acquisition system for creating and managing one or more predictive models that predict values of pricing data variables in distributed systems, said system comprising:

means for dynamically constructing a predictive model of pricing data, based on one or more non-static pricing data variable values, on both a manager system and a disconnectable managed pricing-data system; and

manager system means for processing requests for pricing data values based on the predictive model on the manager system.