# US 2012/0253762 A1 · System and method for computations utilizing optimized earth model representations

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2012/0253762 A1
- Document ID: US20120253762A1
- Status: Published application
- Kind: A1
- Application number: 13/075,329
- Filed: 2011-03-30
- Published: 2012-10-04
- Earliest claimed: 2011-03-30
- Assignees: Chevron U.S.A. Inc.
- Inventors: Raymond Ergas; Oliver Pell; Tamas Nemeth
- Web page: https://patentgrep.com/patent/US20120253762A1

## CPC

- G01V 11/00: Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
- G06T 9/001: {Model-based coding, e.g. wire frame}
- G06T 9/005: {Statistical coding, e.g. Huffman, run length coding}
- G01V 2210/661: Model from sedimentation process modeling, e.g. from first principles
- G01V 2210/665: using geostatistical modeling
- G01V: GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS (means for indicating the location of accidentally buried, e.g. snow-buried, persons A63B29/02)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method and corresponding system is provided for computation utilizing an earth model representation via a computing system having a first processor having access to an earth model dataset. The method includes compressing the earth model dataset at the first processor to generate a look-up table and a set of data indices (i.e., collectively a compressed earth model representation), wherein the look-up table includes quantized data values. By then storing the look-up table in a first level (“fast”) memory, and storing the indices in a second level (“slower,” higher memory capacity) memory, the look-up table and the indices can be accessed to selectively decompress the compressed earth model representation at the first processor such that the computation can be performed efficiently by the first processor.

## Claims

### Claim 1

A method for computation utilizing an earth model representation, the method being executed via a computing system having a first processor having access to an earth model dataset, the first processor being operatively in communication with a first level memory and a second level memory, the second level memory being slower but with a higher memory capacity than the first level memory, the method comprising:
compressing the earth model dataset via the first processor to generate a compressed earth model representation, the compressed earth model representation comprising a plurality of data indices and a look-up table comprising quantized data values;
storing the look-up table in the first level memory;
storing the indices in the second level memory; and
accessing the look-up table from the first level memory and the indices from the second level memory to selectively decompress the compressed earth model representation at the first processor to enable the computation by the first processor.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the compressing step comprises non-uniformly quantizing the earth model dataset to generate at least one of the data values.

### Claim 3 (depends on 2)

The method according to claim 2, wherein the non-uniformly quantizing step comprises using a cubic quantization.

### Claim 4 (depends on 2)

The method according to claim 2, wherein the non-uniformly quantizing step comprises using an adaptive quantization.

### Claim 5 (depends on 2)

The method according to claim 2, wherein the non-uniformly quantizing step comprises using a guided adaptive quantization.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the compressing step comprises uniformly quantizing the earth model dataset to generate at least one of the data values.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises a plurality of earth model parameters, and wherein the compressing step comprises compressing at least two of the earth model parameters to assure consistency with constraints.

### Claim 8 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises a plurality of earth model parameters represented by a plurality of vector values, and wherein the compressing step comprises compressing the vector values in parallel.

### Claim 9 (depends on 1)

The method according to claim 1, wherein the compressed earth model representation comprises a plurality of vector values, and wherein the vector values are decompressed in parallel.

### Claim 10 (depends on 1)

The method according to claim 1, wherein the compressing step further comprises dithering the quantized data values of the look-up table to improve accuracy in the computations.

### Claim 11 (depends on 1)

The method according to claim 1, wherein the first processor comprises a central processing unit (CPU) and the first level memory comprises a cache memory.

### Claim 12 (depends on 1)

The method according to claim 1, further comprising:
a second processor in communication with the first processor, the second processor having a first level memory and a second level memory, the second level memory of the second processor being slower but with a higher memory capacity than the first level memory of the second processor; and
wherein the look-up table is instead stored in the first level memory of the second processor, the indices are instead stored in the second level memory of the second processor, and wherein the look-up table and the indices are instead accessed from the first level and second level memories of the second processor, respectively, to selectively decompress the compressed earth model representation at the second processor instead of the first processor and to enable the computation by the second processor instead of the first processor.

### Claim 13 (depends on 12)

The method according to claim 12, wherein the second processor comprises a graphics processing unit (GPU) and the first level memory comprises a shared memory.

### Claim 14 (depends on 12)

The method according to claim 12, wherein the second processor comprises a field-programmable gate array (FPGA) and the first level memory comprises a Block RAM (BRAM).

### Claim 15 (depends on 1)

The method according to claim 1, wherein the data values of the look-up table comprise one or more of derived, scalar and vector values.

### Claim 16 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises acoustic model parameters.

### Claim 17 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises vertical transverse isotrophy (VTI) model parameters.

### Claim 18 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises tilted transverse isotropy (TTI) model parameters.

### Claim 19 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises variable density TTI model parameters.

### Claim 20 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises elastic model parameters.

### Claim 21 (depends on 1)

The method according to claim 1, wherein the earth model dataset comprises visco-elastic model parameters.

### Claim 22 (depends on 1)

The method according to claim 1, wherein the computation comprises seismic processing.

### Claim 23

A system for computation utilizing an earth model representation, comprising:
a first level memory;
a second level memory being slower but with a higher memory capacity than the first level memory;
a first processor having access to an earth model dataset, the first processor being operatively in communication with the first level memory and the second level memory;
computer readable media accessible by the first processor, the computer readable media comprising computer readable code for:
compressing the earth model dataset at the first processor to generate a compressed earth model representation, the compressed earth model representation comprising a plurality of data indices and a look-up table comprising quantized data values;
storing the look-up table in the first level memory;
storing the indices in the second level memory; and
accessing the look-up table from the first level memory and the indices from the second level memory to selectively decompress the compressed earth model representation at the first processor to enable the computation by the first processor.

### Claim 24 (depends on 23)

The system according to claim 23, wherein the processor comprises a central processing unit (CPU) and the first level memory comprises a level-one (L1) cache memory.

### Claim 25 (depends on 23)

The system according to claim 23, further comprising:
a second processor in communication with the first processor, the second processor having a first level memory and a second level memory, the second level memory of the second processor being slower but with a higher memory capacity than the first level memory of the first processor; and
wherein the computer readable media is accessible by one or both of the first and second processors, and wherein the computer readable media further comprises computer readable code for:
storing the look-up table in the first level memory of the second processor instead of the first level memory of the first processor;
storing the indices in the second level memory of the second processor instead of the second level memory of the first processor; and
accessing the look-up table and the indices from the first level and second level memories of the second processor, respectively, to selectively decompress the compressed earth model representation at the second processor instead of the first processor to enable the computation by the second processor instead of the first processor.

### Claim 26 (depends on 25)

The system according to claim 25, wherein the second processor comprises a graphics processing unit (GPU) and the first level memory comprises a shared memory.

### Claim 27 (depends on 25)

The system according to claim 25, wherein the second processor comprises a field-programmable gate array (FPGA) and the first level memory comprises a Block RAM (BRAM).

### Claim 28

An article of manufacture comprising a computer readable medium having a computer readable code embodied therein, the computer readable code adapted to be executed to implement a method for computation utilizing an earth model representation, the method comprising:
compressing an earth model dataset at the first processor to generate a compressed earth model representation, the compressed earth model representation comprising a plurality of data indices and a look-up table comprising quantized data values;
storing the look-up table in a first level memory;
storing the compressed earth model representation in a second level memory, the second level memory being slower but with a higher memory capacity than the first level memory; and
accessing the look-up table from the first level memory and the indices from the second level memory to selectively decompress the compressed earth model representation at the first processor to enable the computation by the first processor.

### Claim 29 (depends on 28)

The article of manufacture according to claim 28, wherein the method further comprises:
storing the look-up table in a first level memory of a second processor instead of the first level memory of the first processor;
storing the indices in a second level memory of the second processor instead of the second level memory of the first processor, second level memory of the second processor being slower but with a higher memory capacity than the first level memory of the second processor; and
accessing the look-up table and the indices from the first level and second level memories of the second processor, respectively, to selectively decompress the compressed earth model representation at the second processor instead of the first processor to enable the computation by the second processor instead of the first processor.
