Patent US8150785

Multi-modal green computing fusion using problem analytics
[+] Show Details

US8150785

1. A system for processing a computational problem for a computing center having a plurality of computing resources, comprising: (5) (4)

a computer system having a processor and memory, and including:

a problem analytics system for generating a characterized problem description from an inputted computational problem using at least one of a keyword nonparametric inverse indexing or a nonparametric data structure analysis;

a system for selecting a model that resembles the characterized problem description by comparing the characterized problem description with a set of models; and

a system for generating a computing solution for running the inputted computational problem on the computing center based on a selected model and a set of environment factors associated with the computing center.

2. The system of claim 1, wherein the system for selecting the model utilizes a pattern matching algorithm. (1) (0)

4. The system of claim 1, wherein the system for selecting the model utilizes a decision algorithm based on metadata. (0)

5. The system of claim 1, wherein the computing resources of the computing center are selected from a group consisting of: a super computer, a DNA computer, a quantum computer, and a grid computer. (0)

6. The system of claim 1, further comprising an off-line model training and evaluation system. (0)

7. The system of claim 1, wherein the system for generating a computing solution includes a neural network. (1) (0)

9. A computer program product stored on a computer readable storage medium for processing a computational problem for a computing center having a plurality of computing resources, comprising: (5) (3)

program code for generating a characterized problem description from an inputted computational problem using at least one of a keyword nonparametric inverse indexing or a nonparametric data structure analysis;

program code for selecting a model that resembles the characterized problem description by comparing the characterized problem description with a set of models; and

program code for generating a computing solution for running the inputted computational problem on the computing center based on a selected model and a set of environment factors associated with the computing center.

10. The computer program product of claim 9, wherein the program code for selecting the model utilizes a pattern matching algorithm. (1) (0)

12. The computer program product of claim 9, wherein the program code for selecting the model utilizes a decision algorithm based on metadata. (0)

13. The computer program product of claim 9, wherein the computing resources of the computing center are selected from a group consisting of: a super computer, a DNA computer, a quantum computer, and a grid computer. (0)

14. The computer program product of claim 9, further comprising program code for providing off-line model training and evaluation. (0)

15. The computer program product of claim 9, wherein the program code for generating a computing solution includes a neural network. (1) (0)

17. A method for processing a computational problem for a computing center having a plurality of computing resources, comprising: (5) (4)

utilizing a computer system having a processor and a memory to perform:

generating a characterized problem description from an inputted computational problem using at least one of a keyword nonparametric inverse indexing or a nonparametric data structure analysis;

selecting a model that resembles the characterized problem description by comparing the characterized problem description with a set of models; and

generating a computing solution for running the inputted computational problem on the computing center based on a selected model and a set of environment factors associated with the computing center.

18. The method of claim 17, wherein selecting the model utilizes a pattern matching algorithm. (1) (0)

20. The method of claim 17, wherein selecting the model utilizes a decision algorithm based on metadata. (0)

21. The method of claim 17, wherein the computing resources of the computing center are selected from a group consisting of: a super computer, a DNA computer, a quantum computer, and a grid computer. (0)

22. The method of claim 17, further comprising providing off-line model training and evaluation. (0)

23. The method of claim 17, wherein generating a computing solution includes utilizing a neural network. (1) (0)

View Abstract and Specification Size

PDF with Images and Document Face >

Full Text Publication >

Patent US8150785

Multi-modal green computing fusion using problem analytics

Patent Matrix® Search