ASIC Boolean Valuation for Sparse Search Spaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Boolean valuation techniques are inefficient in processing large rule sets and identifying satisfied conjunctions, particularly in scalable and sparse data structures, which limits their effectiveness in search operations over complex search spaces.
Innovation Solution
A system utilizing a three-layer ASIC architecture with a bus of parallel conductors and AND-gates, where input literals are encoded as bit vectors and processed to identify satisfied clauses by incrementing state vectors and comparing them to target vectors, enabling efficient Boolean valuation and search result generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional Boolean valuation techniques (heuristic approaches, BDDs, tries, hypergraph-based approaches) are used, then the implementation is straightforward with existing data structures, but the processing efficiency deteriorates when handling large rule sets and sparse data structures
Solution Approach 1:
The patent replaces conventional software-based Boolean valuation techniques (heuristic approaches, BDDs, tries, hypergraph-based approaches) with a hardware-based ASIC system. The ASIC uses parallel AND-gates to simultaneously evaluate multiple conjunctions, substituting sequential software processing with parallel hardware computation, thereby dramatically improving processing efficiency for large rule sets
Solution Approach 2:
The patent segments the Boolean valuation process into distinct hardware components: input literal encoding units, parallel AND-gate arrays for conjunction evaluation, state vector incrementers, and comparator units. Each component handles a specific aspect of the valuation process, allowing independent optimization and simultaneous operation to improve overall productivity
2Adaptability or versatility
If the rule set scale increases, then the search space coverage improves, but the processing time and computational resources worsen
Solution Approach 1:
The patent transitions from sequential software-based processing to parallel hardware-based processing by mapping the Boolean valuation problem onto a three-dimensional ASIC architecture. The system uses parallel AND-gates arranged in a spatial configuration that enables simultaneous evaluation of multiple conjunctions across different dimensions of the search space, achieving O(1) time complexity for rule set evaluation regardless of scale
Solution Approach 2:
The system pre-encodes input literals into bit vectors and pre-configures the AND-gate connections according to the rule set structure before processing begins. This preliminary setup allows the ASIC to immediately begin parallel evaluation when queries are received, eliminating runtime computation overhead and reducing processing time for large rule sets
Data Source
AI summary
Various technologies for performing Boolean valuation of a vector of input literals over a clause set representing a search space are described herein. For each of a plurality of literals in a literal vector, state vector entries that correspond to clauses in which the literal appears are incremented. Clauses are output as results based upon determining that state vector entries corresponding to the clauses meet or exceed target vector entries corresponding to the clauses.