Content-Associative Sieve for Lossless Data Reduction at Scale

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data compression methods are inefficient in handling large and extremely large datasets, as they can only exploit redundancy within a local window and are compute-intensive, leading to throttling by storage access bandwidth limitations, and are not suited for random access or high-speed data ingestion and retrieval.

Innovation Solution

The Data Distillation process factorizes input data into Prime Data Elements and Derivative Elements, using a content-associative sieve to identify and reconstitute data chunks, allowing for lossless reduction and efficient retrieval by referencing Prime Data Elements and a reconstitution program, enabling data reduction across large datasets.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If traditional compression methods are used to reduce data footprint, then data storage efficiency improves, but processing speed and ingestion rate deteriorate due to compute-intensive operations

Engineering Contradiction:
Improvedata footprintVSAvoidprocessing speed
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent segments data into fixed-size chunks and processes them independently through the content-associative sieve. This segmentation allows parallel processing of multiple chunks simultaneously, improving throughput while maintaining compression effectiveness. Each chunk is handled as a discrete unit, enabling pipelining and reducing sequential processing bottlenecks.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-processing data chunks to extract features and create representations before compression. The content-associative sieve pre-computes hashes and organizes data structures in advance, so when compression is needed, the system can quickly lookup and reference existing prime data elements without intensive real-time computation.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If redundancy exploitation is limited to local windows, then processing efficiency improves, but data reduction capability deteriorates due to inability to find global redundancies

Engineering Contradiction:
Improveprocessing efficiencyVSAvoiddata reduction capability
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent transitions from local window-based compression to a global content-associative space by hashing data chunks and organizing them in a sieve structure. This dimensional change allows the system to search across the entire dataset for redundant content rather than being constrained to local windows, enabling discovery of global redundancies while maintaining efficient lookup through hash-based indexing.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The content-associative sieve acts as an intermediary structure between raw data and compressed output. It hashes and indexes prime data elements, creating a mediator layer that enables efficient global redundancy detection. The sieve structure with its hash functions and bucket organization serves as the intermediary that connects input chunks to matching prime elements, allowing global search without sacrificing processing efficiency.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Quantity of substance

If compute-intensive compression algorithms are used, then data reduction ratio improves, but storage access bandwidth utilization deteriorates due to throttling

Engineering Contradiction:
Improvedata reduction ratioVSAvoidbandwidth utilization
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The patent replaces compute-intensive mechanical compression algorithms with a lookup-based system using content-associative sieves. Instead of performing intensive computations to find redundancies, the system uses hash functions and direct table lookups to identify prime data elements. This substitution of computational mechanics with data structure-based mechanics dramatically reduces processing load and improves bandwidth utilization.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the parameter of redundancy detection from computational complexity to data structure organization. By organizing prime data elements in a content-associative sieve with specific hash function parameters and bucket structures, the system achieves high data reduction ratios through efficient parameterized lookup rather than intensive computation, thereby improving bandwidth efficiency.

Inventive Principle:
Principle #35Parameter changes

4Quantity of substance

If data is compressed using existing methods, then storage density improves, but random access capability deteriorates due to inability to efficiently retrieve specific chunks

Engineering Contradiction:
Improvestorage densityVSAvoidrandom access capability
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The patent creates content-associative copies and references of prime data elements in the sieve structure. When data is compressed, instead of storing only compressed bits, the system maintains reference copies in the content-associative sieve that enable rapid retrieval. These copied references act as indexes that allow random access to specific chunks without decompressing entire data streams, maintaining storage density while enabling efficient access.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3238344B1Lossless reduction of data by deriving data from prime data elements resident in a content-associative sieve
Publication Date: 2022.03.09 ASCAVA INC
  • EP3238344B1 patent drawingFigure 1A
  • EP3238344B1 patent drawingFigure 1B
  • EP3238344B1 patent drawingFigure 1C

AI summary

This disclosure relates to lossless data reduction on large and extremely large datasets while providing high rates of data ingestion and data retrieval. Some embodiments can generate a losslessly reduced representation of a data chunk, wherein the losslessly reduced representation includes a reference to one or more prime data elements stored in a content associative sieve, and optionally a description of a reconstitution program which, when applied to the one or more prime data elements results in the data chunk. The losslessly reduced representation of the data chunk can then be communicated to another computer over a network.