Division of processing between systems based on external factors

• US 9,483,811 B2
• Filed: 01/06/2014
• Issued: 11/01/2016
• Est. Priority Date: 01/06/2014
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Ahmed, Haroon, Hill, David Paul, Bruckner, Robert M., McKay, Patrick, Campbell, David G.
• Primary Examiner(s): Tryder, Gregory J
• Assistant Examiner(s): Le, Michael
• Application Number:
  US14/148,421
• Publication Number:
  US 20150193901A1
• Time in Patent Office: 1,030 Days
• Field of Search: 345/502, 345/505, 345/506, 345/718
• US Class Current: 1/1
• CPC Class Codes: G06F 2209/509 Offload
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining

Abstract

A method includes acts for rendering, on a data processing system, a result derived from a set of data by performing data processing across a first data processing system and a second data processing system. The amount of processing performed by the second data processing system can be dynamically adjusted depending on factors affecting the second data processing system. The first data processing system receives information defining how the result will be rendered at the second data processing system. The first data processing system receives information indicating factors affecting the second data processing system. The first data processing system dynamically allocates the needed data processing between the first data processing system and the second data processing system, based on factors affecting the second data processing system.

Claims

1. A method of rendering, on a second data processing system, a result derived from a set of data by performing data processing across a first data processing system and the second data processing system, where the amount of processing performed by the second data processing system is dynamically adjusted depending on a machine learning model, the method comprising:

the first data processing system receiving information defining how a result associated with a data processing task will be rendered at the second data processing system;

the first data processing system identifying a machine learning model that models user behavior at the second data processing system in connection with operation of the data processing task; and

the first data processing system dynamically allocating a first portion of needed data processing for the data processing task to the first data processing system and allocating a second portion of the needed data processing to the second data processing system, based on an amount of modeled user behavior at the second data processing system in connection with operation of the data processing task that is modeled by the machine learning model, including;

based at least on the amount of modeled user behavior being below a defined threshold, allocating the first portion of the needed data processing to the first data processing system as a larger portion of the needed data processing than the second portion of the needed data processing that is allocated to the second data processing system;

or

based at least on the amount of modeled user behavior being above the defined threshold, allocating the first portion of the needed data processing to the first data processing system as a smaller portion of the needed data processing than the second portion of the needed data processing that is allocated to the second data processing system,

wherein the first data processing system maintains the machine learning model when the amount of modeled user behavior is below the defined threshold, and wherein the first data processing system installs the machine learning model at the second data processing system when the amount of modeled user behavior reaches the defined threshold.

2. The method of claim 1, wherein the first data processing system is a server and the second data processing system is a client.

3. The method of claim 1, further comprising receiving information defining a fault domain for the second data processing system.

4. The method of claim 1, further comprising receiving information indicating network quality between the second data processing system and the first data processing system.

5. The method of claim 1, wherein further comprising receiving information indicating a degree of integration between the second data processing system and other entities.

6. The method of claim 1, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed using a continuous calibration of one or more factors affecting the second data processing system.

7. The method of claim 1, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed each time a session is initialized between the first data processing system and the second data processing system.

8. The method of claim 1, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed on a change in one or more factors affecting the second data processing system.

9. The method of claim 1, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed for each request from the second data processing system to the first data processing system.

10. A system for rendering, on a second data processing system, a result derived from a set of data by performing data processing across a first data processing system and the second data processing system, where the amount of processing performed by the second data processing system is dynamically adjusted depending on a machine learning model, the system comprising:

one or more processors;

a network communications device; and

one or more computer readable media, wherein the one or more computer readable media comprise computer executable instructions that are executable by at least one of the one or more processors to configure the system to perform at least the following;

receive, over the network communications device, information defining how a result associated with a data processing task will be rendered at the second data processing system;

identify a machine learning factor comprising a machine learning model that models user behavior at the second data processing system in connection with operation of the data processing task; and

dynamically allocate a first portion of needed data processing for the data processing task to the first data processing system and allocating a second portion of the needed data processing to the second data processing system, based on an amount of modeled user behavior at the second data processing system in connection with operation of the data processing task that is modeled by the machine learning model, including;

based at least on the amount of modeled user behavior being below a defined threshold, allocating the first portion of the needed data processing to the first data processing system as a larger portion of the needed data processing than the second portion of the needed data processing that is allocated to the second data processing system;

or

based at least on the amount of modeled user behavior being above the defined threshold, allocating the first portion of the needed data processing to the first data processing system as a smaller portion of the needed data processing than the second portion of the needed data processing that is allocated to the second data processing system,

wherein the first data processing system maintains the machine learning model when the amount of modeled user behavior is below the defined threshold, and wherein the first data processing system installs the machine learning model at the second data processing system when the amount of modeled user behavior reaches the defined threshold.

11. The system of claim 10, wherein the allocation is also based on a fault domain for the second data processing system.

12. The system of claim 10, wherein the allocation is also based on network quality between the second data processing system and the first data processing system.

13. The system of claim 10, wherein the allocation is also based on a degree of integration between the second data processing system and other entities.

14. The system of claim 10, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed using a continuous calibration of one or more factors affecting the second data processing system.

15. The system of claim 10, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed each time a session is initialized between the first data processing system and the second data processing system.

16. The system of claim 10, wherein dynamically allocating first and second portions of the needed data processing between the first data processing system and the second data processing system is performed on a change in one or more factors affecting the second data processing system.

17. The system of claim 10, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed for each request from the second data processing system to the first data processing system.

18. A physical computer readable storage device comprising computer executable instructions that are executable by at least one processor of a first data processing system to configure the first data processing system to perform at least the following:

receive information defining how a result associated with a data processing task will be rendered at the second data processing system;

identify a machine learning factor comprising a machine learning model that models user behavior at the second data processing system in connection with operation of the data processing task; and

dynamically allocate a first portion of needed data processing for the data processing task to the first data processing system and allocating a second portion of the needed data processing to the second data processing system, based on an amount of modeled user behavior at the second data processing system in connection with operation of the data processing task that is modeled by the machine learning model, including;

based at least on the amount of modeled user behavior being below a defined threshold, allocating the first portion of the needed data processing to the first data processing system as a larger portion of the needed data processing than the second portion of the needed data processing that is allocated to the second data processing system;

or

based at least on the amount of modeled user behavior being above the defined threshold, allocating the first portion of the needed data processing to the first data processing system as a smaller portion of the needed data processing than the second portion of the needed data processing that is allocated to the second data processing system,

wherein the first data processing system maintains the machine learning model when the amount of modeled user behavior is below the defined threshold, and wherein the first data processing system installs the machine learning model at the second data processing system when the amount of modeled user behavior reaches the defined threshold.

19. The physical computer readable storage device of claim 18, wherein the allocation is also based on a fault domain for the second data processing system.

20. The physical computer readable storage device of claim 18, wherein the allocation is also based on network quality between the second data processing system and the first data processing system.

21. The physical computer readable storage device of claim 18, wherein the allocation is also based on a degree of integration between the second data processing system and other entities.

22. The physical computer readable storage device of claim 18, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed using a continuous calibration of one or more factors affecting the second data processing system.

23. The physical computer readable storage device of claim 18, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed each time a session is initialized between the first data processing system and the second data processing system.

24. The physical computer readable storage device of claim 18, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed on a change in one or more factors affecting the second data processing system.

25. The physical computer readable storage device of claim 18, wherein dynamically allocating the first and second portions of the needed data processing between the first data processing system and the second data processing system is performed for each request from the second data processing system to the first data processing system.