Coprocessor Data Conversion Using Pattern Frequency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing data conversion technologies face inefficiencies due to high power consumption and reduced throughput when handling numerous text characters across multiple languages, as they process each input pattern individually, leading to increased power usage and reduced performance.

Innovation Solution

A coprocessor system that utilizes statistical occurrence of data patterns to convert frequently occurring patterns in a single conversion cycle, employing a converter engine with a parser and converter to determine high likelihood patterns and optimize data conversion by grouping patterns into high and low frequency categories, allowing for efficient conversion of high frequency patterns in one cycle and breaking down low frequency patterns into multiple cycles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If each input pattern is processed individually, then data conversion accuracy is maintained, but power consumption increases and throughput decreases

Engineering Contradiction:

ImprovethroughputVSAvoidpower consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent merges multiple frequently occurring input patterns into a single conversion cycle by identifying common byte sequence patterns. The converter engine processes high-frequency patterns collectively rather than individually, combining multiple conversion operations into one unified processing step, which reduces power consumption and increases throughput.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The system performs preliminary analysis to determine the frequency of occurrence of different data patterns before conversion. By pre-identifying high-frequency patterns and preparing conversion rules in advance, the system optimizes the conversion process to handle these patterns more efficiently, reducing both power consumption and processing time.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If all data patterns are supported, then data conversion versatility is maintained, but device complexity and power requirements increase

Engineering Contradiction:

Improvedata pattern supportVSAvoidlogic requirements

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments data patterns into high-frequency and low-frequency categories based on their occurrence probability. This segmentation allows the converter engine to apply different processing strategies: high-frequency patterns use optimized conversion rules, while low-frequency patterns follow standard conversion procedures, thereby reducing overall device complexity while maintaining versatility.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system applies different levels of processing complexity to different data patterns based on their frequency characteristics. High-frequency patterns receive specialized optimized handling with simplified logic, while low-frequency patterns use comprehensive conversion rules. This local differentiation reduces overall device complexity while preserving full adaptability.

Inventive Principle:

Principle #3Local quality

3Productivity

If standard conversion processing is used for all patterns, then conversion accuracy is maintained, but processing time increases

Engineering Contradiction:

Improveconversion speedVSAvoidconversion accuracy

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The converter engine dynamically adjusts its processing approach based on the identified data pattern frequency. For high-frequency patterns, it applies optimized conversion rules that prioritize speed while maintaining accuracy. For low-frequency patterns, it uses standard comprehensive processing. This dynamic adaptation allows the system to maximize conversion speed without sacrificing accuracy.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes processing parameters such as conversion cycle duration and pattern matching depth based on the frequency characteristics of input data. High-frequency patterns are processed with optimized parameters that reduce processing time while maintaining accuracy standards, ensuring improved conversion speed without compromising precision.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10915547B2Optimizing data conversion using pattern frequency

Publication Date: 2021.02.09 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments of the present invention provide systems and methods for increasing the efficiency of data conversion in a coprocessor by using the statistical occurrence of data patterns to convert frequently occurring data patterns in one conversion cycle. In one embodiment, a coprocessor system is disclosed containing a converter engine, which includes a parser and a converter, an input buffer, and a result store. The input buffer is configured to transfer a set of source data to the converter engine, which converts the source data from first code format to a second code format, and sends the converted source data to the result store.