Bloom Filter Entry Removal Using LSB Toggling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Bloom filters, while memory-efficient for determining set membership, lack the ability to remove entries without impacting other entries or sacrificing memory efficiency, as changing bits from one to zero can compromise the integrity of the filter.

Innovation Solution

The implementation of an iterative process using an entry function with an iteration value to add or remove entries from the Bloom filter, where even iteration values indicate presence and odd values indicate absence, allowing for efficient addition and removal of entries without compromising the filter's integrity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If counting filters with n-bit counters are used to remove entries from the Bloom filter, then the ability to remove entries is improved, but memory efficiency deteriorates as each segment comprises many more bit values

Engineering Contradiction:
Improveability to remove entriesVSAvoidmemory size
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent changes the parameter representation from storing full counter values (n-bit counters) to storing only the least significant bit (LSB). This parameter transformation allows the same memory structure to support both addition and removal operations while maintaining memory efficiency. The LSB acts as a toggle state that indicates whether an element is present or removed, eliminating the need for multi-bit counters.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If Bloom filter bits are changed from one to zero to remove entries, then entry removal is enabled, but the integrity of the Bloom filter is compromised

Engineering Contradiction:
Improveentry removal capabilityVSAvoidfilter integrity
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces dynamic iteration values that change with each add or remove operation. Instead of statically setting bits to zero, the system dynamically toggles the LSB based on the current iteration value. This dynamic approach ensures that removal operations are reversible and maintain filter integrity, as the same bit can be toggled back to its original state when needed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses feedback from the iteration value to control bit toggling. Each operation increments the iteration value, and the LSB of this increment determines whether to toggle the filter bit. This feedback mechanism ensures that removal operations are coordinated with the overall filter state, preventing integrity compromise while enabling entry removal.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS10579831B2Verification of data set components using digitally signed probabilistic data structures
Publication Date: 2020.03.03 AMAZON TECH INC
  • US10579831B2 patent drawing
  • US10579831B2 patent drawing
  • US10579831B2 patent drawing

AI summary

A computer system encodes a plurality of components of a data set into a probabilistic data structure and digitally signs the probabilistic data structure. The computer system provides the digital signature for the probabilistic data structure and the probabilistic data structure to various entities. An entity can verify an individual component of the data set within the probabilistic data structure by verifying the individual component against the probabilistic data structure and the digital signature of the probabilistic data structure.