Techniques for tracking resource usage statistics per transaction across multiple layers of protocols

• US 9,686,152 B2
• Filed: 01/27/2012
• Issued: 06/20/2017
• Est. Priority Date: 01/27/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Tejerina, David Nunez, Mathur, Siddhartha, Ozhan, Ozan, Kolluru, Raghu, Hamza, Adrian, Kleewein, James
• Primary Examiner(s): Tran, Jimmy H
• Application Number:
  US13/360,460
• Publication Number:
  US 20130198363A1
• Time in Patent Office: 1,971 Days
• Field of Search: 709223, 709224, 714 43
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: H04L 43/091 Measuring contribution of individual network components to actual service level
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Techniques to track resource usage statistics per transaction across multiple layers of protocols and across multiple threads, processes and/or devices are disclosed. In one embodiment, for example, a technique may comprise assigning an activity context to a request at the beginning of a first stage, where the activity context has an initial set of properties. The values of the properties may be assigned to the properties in the initial set during the first stage. The value of a property may be stored on a data store local to the first stage. The activity context may be transferred to a second stage when the request begins the second stage. The transferred activity context may include a property from the initial set of properties. The stored values may be analyzed to determine a resource usage statistic. Other embodiments are described and claimed.

Claims

1. A computer-implemented method, comprising:

receiving a request for data, services, or both data and services;

processing the request across a plurality of stages, beginning with a first stage;

assigning an activity context to the request at the beginning of the first stage, wherein the activity context includes an initial set of a plurality of properties that describe the request, wherein the plurality of properties includes;

an activity identifier that identifies the request;

a tenant identifier that identifies a domain name of a tenant receiving services;

a component identifier that identifies a component that handles the request; and

an action identifier that identifies an action of the request;

assigning values to the properties in the initial set during the first stage;

storing the assigned value to the properties on a data store local to the first stage;

transferring the activity context to a second stage when the request begins the second stage;

collecting the stored values;

generating a log file from the stored values;

analyzing the log file to determine a resource usage statistic; and

using the resource usage statistic to perform at least one of;

adjust a throttling algorithm;

determine a peak usage time;

determine a response time;

determine a relative resource usage of a first client-initiated action and a second client-initiated action; and

debug a resource performance problem.

2. The method of claim 1, wherein a stage comprises at least one of a thread or a process.

3. The method of claim 1, further comprising:

determining at least one of;

a number of processing unit cycles;

a number of read operations;

a number of write operations;

a number of database accesses;

a number of times a protocol is used;

a number of requests from an application;

a number of requests from a client;

a number of requests from a user;

a latency time;

a total activity time; and

a peak request time.

4. The method of claim 1, further comprising transferring a subset of the plurality of properties of the activity context to the second stage, the subset comprising fewer properties than all of the plurality of properties in the activity context.

5. The method of claim 1, wherein a property comprises at least one of:

the activity identifier;

a user identifier;

an email address;

an authentication type;

an authentication token;

the tenant identifier;

a tenant type;

a component;

a component instance;

a feature;

a protocol of the request;

client information;

the action identifier;

a resource usage metric;

metadata; and

a client version.

15. The method of claim 4 further comprising:

assigning values to the subset of the properties of the activity context in the second stage; and

storing the assigned value to the properties on a data store local to the second stage.

16. The method of claim 1 wherein:

the receiving of the request for data, services, or both data and services is performed by a web services server;

the request for data, services, or both data and services originated from a user device being operated by a user to access the web services server; and

the initial set of the plurality of properties that describe the request includes a unique identifier of the user.

17. The method of claim 1, further comprising a workload manager to analyze the log file to determine a resource usage statistic.

6. An article comprising a computer readable storage memory unit comprising instructions that when executed cause a system to:

receive one or more requests for at least one of data and services corresponding to an application;

process the one or more requests across a plurality of stages, beginning with a first stage;

assign an activity context to each of the one or more requests at the beginning of the first stage, wherein the activity context includes an initial set of a plurality of properties that describe the each request, wherein the plurality of properties includes;

an activity identifier that uniquely identifies that request;

a tenant identifier that identifies a domain name of a tenant receiving services; and

an action identifier that identifies an action of the request;

assign values to the properties in the first stage;

store the values assigned to the properties in the first stage on a data store local to the first stage;

transfer the activity context to a second stage when the each request begins the second stage;

analyze the stored values of the requests to generate resource usage statistics for the application; and

using the resource usage statistics to perform at least one of;

adjust a throttling algorithm;

determine a peak usage time;

determine a response time;

determine a relative resource usage of a first client-initiated action and a second client-initiated action; and

debug a resource performance problem.

7. The article of claim 6, the memory unit further comprising instructions that when executed cause the system to:

collect the stored values; and

generate a log file from the stored values.

8. The article of claim 6, the memory unit further comprising instructions that when executed cause the system to determine at least one of:

a number of processing unit cycles;

a number of read operations;

a number of write operations;

a number of database accesses;

a number of times a protocol is used;

a number of requests from the application;

a number of requests from a client;

a number of requests from a user;

a latency time;

a total activity time; and

a peak request time.

9. The article of claim 6, the memory unit further comprising instructions that when executed cause the system to add a property to the set of properties during a stage.

10. The article of claim 6, the memory unit further comprising instructions that when executed cause the system to transfer a subset of the properties of the activity context to the second stage, the subset comprising fewer properties than all of the properties in the activity context.

18. The article of claim 6, the memory unit further comprising instructions that when executed provide a workload manager to analyze the log file.

19. The article of claim 6, wherein:

the receiving of the one or more requests for at least one of data and services is performed by a web services server;

the one or more requests for at least one of data and services originated from a user device being operated by a user to access the web services server; and

the initial set of the plurality of properties that describe the request includes a unique identifier of the user.

11. An apparatus, comprising:

a processing unit;

a memory to store a data store;

a resource handler executing on the processing unit to;

receive a request for data, services, or both data and services;

process the request across a plurality of stages, beginning with a first stage comprising a thread or a process;

assign an activity context to the request at the beginning of the first stage, wherein the activity context includes an initial set of a plurality of properties that describe the request, wherein the plurality of properties includes;

an activity identifier that uniquely identifies the request;

a tenant identifier that identifies a domain name of a tenant receiving services;

a component identifier that identifies a component that handles the request; and

an action identifier that identifies an action of the request;

assign values to the properties in the initial set during the first stage;

store the values assigned to the properties in the first stage a value of a property on the data store;

transfer the activity context to a second stage when the request begins the second stage, the second stage comprising another thread or another process;

analyze the stored values to determine a resource usage statistic; and

use the resource usage statistic to perform at least one of;

adjust a throttling algorithm;

determine a peak usage time;

determine a response time;

determine a relative resource usage of a first client-initiated action and

a second client-initiated action; and

debug a resource performance problem.

12. The apparatus of claim 11, further comprising a logger executing on the processing unit to:

collect the stored values;

generate a log file from the stored values; and

store the log file on the data store.

13. The apparatus of claim 12, further comprising a workload manager executing on the processing unit to analyze the log file to determine a resource usage statistic.

14. The apparatus of claim 11, wherein a property is added to the initial set of the plurality of properties during a stage.

20. The apparatus of claim 11, wherein:

the receiving of the request for data, services, or both data and services is performed by a web services server;

the request for data, services, or both data and services originated from a user device being operated by a user to access the web services server; and

the initial set of the plurality of properties that describe the request includes a unique identifier of the user.