Optimized performance counter monitoring

• US 7,698,417 B2
• Filed: 06/15/2005
• Issued: 04/13/2010
• Est. Priority Date: 06/15/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Rizzi, Lorenzo, Dilillo, Gerardo
• Primary Examiner(s): Etienne; Ario
• Assistant Examiner(s): Todd; Gregory G
• Application Number:
  US11/152,941
• Publication Number:
  US 20060287738A1
• Time in Patent Office: 1,763 Days
• Field of Search: 709/223, 709/224, 709/232, 709/233, 709/235, 700 32- 36, 702/62, 702/182, 702186-189, 715/734, 715/736
• US Class Current: 709/224
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/81 Threshold

Abstract

In a management system, computing devices collect sample performance values that describe utilization or usage of resources or processes. The sample performance values may be in the form of performance data or performance counters. Sample performance values are either sent or not sent to a management server, based on a delta value of a current collected sample performance value and a last current collected sample value. Plots may be constructed based on actual received sample performance values and interpolated, or estimated sample performance values based on a known frequency at which the sample performance values are collected.

Claims

1. A method comprising:

configuring a plurality of computing devices to communicate with a management server via a network having limited bandwidth, wherein each computing device includes a computing device processor for executing operations on the computing device, wherein the management server includes a server processor for executing operations on the management server;

activating a management service on the management server by the server processor;

activating management agents on each of the plurality of computing devices by the corresponding computing device processors;

monitoring computing device resources on each computing device by the corresponding management agent active on that computing device;

sending by each management agent to the management service a first sample performance value representative of a performance of the computing device resources taken at a first point in time, whereby an initial sample performance value and initial point in time is established at the management server for each computing device;

taking, by each management agent on each computing device, succeeding sample performance values for that computing device at succeeding points in time according to a defined frequency known to each computing device and the management server;

determining, by each management agent on each computing device, whether a particular succeeding sample performance value differs from an immediately preceding sample performance value by a predetermined percentage, wherein the predetermined percentage defines a range greater than zero defined by a positive value greater than the immediately preceding sample performance value and a negative value less than the immediately preceding sample performance value;

sending, by each management agent on each computing device, the particular succeeding sample performance value and a corresponding time value to the management server when the particular succeeding sample performance value differs from the immediately preceding sample performance value by more than the predetermined percentage, and not sending the particular succeeding sample performance value when the particular succeeding sample performance value does not differ from the immediately preceding sample performance value by more than the predetermined percentage, so that bandwidth usage of the network by the management agents is minimized;

receiving, by the management service from each of the computing devices, the first sample performance values taken at the first points in time, and the particular succeeding sample performance values which differed from the immediately preceding sample performance values by more than the predetermined percentage;

plotting, by the management service, the sample performance values received from the computing devices on corresponding plots relative to the times at which the sample performance values were taken, wherein for succeeding points in time according to the defined frequency at which the management server does not receive succeeding sample performance values, the management server plots points in the plots corresponding with values of a most recently received sample performance value for each corresponding computing device.

2. The method of claim 1, wherein the sample performance values are directed to utilization of the computing device processor of each computing device.

3. The method of claim 1, wherein the sample performance values are directed to utilization of a processor on each of the computing devices.

4. The method of claim 1, further comprising:

prior to activating the management agents on each of the plurality of computing devices, receiving the management agents at each of the computing devices from the management server.

5. The method of claim 1, wherein the sample performance values are directed to utilization of a memory on each of the computing devices.

6. The method of claim 1, wherein the sample performance values are directed to performance of one or more processes running on each of the computing devices.

7. One or more computer-readable storage media comprising computer-executable instructions implemented by the processors on the computing devices and the management server for performing the method as recited in claim 1.

8. A computing device comprising:

a processor coupled to a memory; and

a management agent accessed and controlled by the processor, the management agent configured to communicate with a management server and provide the management server with selective sample performance values collected at the computing device indicative of performance of the computing device,

wherein the management agent is configured to provide the management server with a first sample performance value taken at a first time, and take a plurality of succeeding sample performance values indicative of performance of the computing device at corresponding succeeding times based on a defined frequency at which the succeeding sample performance values are taken,

wherein the defined frequency is fixed and known to the computing device and the management server,

wherein, for each particular succeeding sample performance value taken, the management agent is configured to determine whether the particular succeeding sample performance value is within a delta value determined based on an immediately preceding sample performance value, wherein the delta value is a defined range greater than zero defined by a positive value greater than the immediately preceding sample performance value and a negative value less than the immediately preceding sample performance value, and send the sample performance value to the management server when the particular succeeding sample performance value is outside the delta value with respect to the immediately preceding sample performance value, and not send the particular succeeding sample performance value when the particular succeeding sample performance value is within the delta value with respect to the immediately

preceding sample performance value,

wherein the management server is configured to create a plot of sample performance values received from the computing device in correlation with corresponding time values according to the defined frequency that is fixed and known to the management server,

wherein the management server creates the plot based on the first sample value and taken at the first time received from the computing device,

wherein for succeeding points in time according to the defined frequency at which the management server does not receive succeeding sample performance values, the management server plots a point in the plot corresponding with a value of a most recently received sample performance value.

9. The computing device of claim 8, wherein the management agent is configured to communicate with a management service interface of the management server, wherein the management agent is updated by updates received from the management server.

10. The computing device of claim 8, wherein the delta value is a specified percentage of plus or minus change over the immediately preceding sample performance value.

11. The computing device of claim 8, wherein the sample performance values include a percentage of utilization of the processor of the computing device at the points in time.

12. A method implemented by a computing device, the method comprising:

connecting the computing device to a network having limited bandwidth, the computing device being configured to periodically communicate with a management server via the network, wherein the computing device includes a computing device processor and a computing device memory for executing operations on the computing device, wherein the management server includes a server processor and a server memory for executing operations on the management server;

determining, by the computing device, at a first point in time, a first sample performance value indicative of performance of the computing device at the first point in time;

sending, by the computing device, the sample performance value to the management server, whereby an initial sample performance value and initial point in time is established at the computing device and the management server;

taking a plurality of succeeding sample performance values indicative of performance of the computing device at corresponding succeeding points in time based on a defined frequency at which the succeeding sample performance values are taken,

wherein the defined frequency is fixed and known to the computing device and the management server,

wherein for each particular succeeding sample performance value taken at each point in time in the defined frequency by the computing device, the method executed by the computing device further comprises;

determining whether the particular succeeding sample performance value is within a predetermined delta value determined based on an immediately preceding sample performance value by determining whether an absolute value of a difference between the particular succeeding sample and the immediately preceding sample is less than the predetermined delta value, wherein the delta value is a defined range greater than zero defined by a positive value greater than the immediately preceding sample performance value and a negative value less than the immediately preceding sample performance value; and

sending the particular succeeding sample performance value to the management server when the particular succeeding sample performance value is greater than the predetermined delta value with respect to the immediately preceding sample performance value, and not sending the particular succeeding sample performance value when the particular succeeding sample performance value is within the predetermined delta value with respect to the immediately preceding sample

performance value,

wherein the corresponding time values are sent with the particular succeeding sample performance values that are sent to the management server, and

wherein the management server creates a plot of sample performance values received from the computing device in correlation with corresponding time values according to the defined frequency that is fixed and known to the management server,

wherein the management server creates the plot based on the initial sample value and the initial point in time received from the computing device,

wherein for succeeding points in time according to the defined frequency at which the management server does not receive succeeding sample performance values, the management server plots a point in the plot corresponding with a value of a most recently received sample performance value.

13. The method of claim 12, wherein the delta value is a specified percentage of plus or minus change over the immediately preceding sample performance value.

14. The method of claim 12, wherein the sample performance values include a percentage of utilization of the one or more processors at the computing device at the points in time.

15. The method of claim 12, wherein the sample performance values are directed to performance of one or more processes running on the computing device.

16. One or more computer-readable storage media comprising computer-executable instructions implemented by the computing device processor for performing the method as recited in claim 12.