Probabilistic Hash Table Deletion via Summary Buckets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large hash tables can be time-consuming and processor-intensive to search through, especially when they exceed local memory limits, necessitating slower access memory and increasing search time, and existing solutions do not efficiently support dynamic insertion and deletion operations.
Innovation Solution
A method involving a summary table with buckets that track deletions using prefix, transit, and signature tables derived from hash values, allowing for efficient determination of entry existence and supporting deletion operations in a probabilistic data structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a large hash table is used to store more data, then the storage capacity is improved, but the search time increases and processor cycles become more intensive
Solution Approach 1:
The patent divides the hash table into multiple levels (primary hash table, secondary hash tables) where the primary table stores only hashes and points to secondary tables that store actual data. This segmentation allows the system to maintain a large storage capacity while reducing search time by avoiding full-table scans and enabling more efficient lookup operations.
Solution Approach 2:
The patent introduces a hierarchical dimension to the traditional flat hash table structure. By adding a second level of hash tables, the system transforms a two-dimensional search problem into a multi-level structure that can be navigated more efficiently, reducing the search space at each level and overall search time.
2Speed
If a large hash table is stored in local memory, then access speed is improved, but the memory requirements exceed local memory limits
Solution Approach 1:
The patent segments the large hash table into a primary hash table that fits in fast local memory (storing only hashes) and secondary hash tables that can be stored in slower, larger capacity memory. This allows the system to maintain fast access for the most frequently accessed data while storing the complete large dataset in more economical memory.
Solution Approach 2:
The patent applies different storage qualities to different parts of the data structure. The primary hash table uses fast local memory for immediate access, while secondary hash tables use slower but larger capacity memory for complete data storage. This local quality differentiation optimizes both speed and memory utilization.
3Quantity of substance
If existing hash table solutions are used, then data storage is achieved, but dynamic insertion and deletion operations are not efficiently supported
Solution Approach 1:
The patent implements a dynamic hash table structure where buckets can be created, deleted, and reconfigured during operation. The system dynamically adjusts the hash table structure based on insertion and deletion operations, maintaining optimal performance for dynamic workloads rather than static data storage.
Solution Approach 2:
The patent enables efficient deletion operations by allowing buckets to be discarded when empty and recovered when needed. This mechanism supports dynamic data structures where entries are frequently added and removed, improving productivity for applications requiring continuous updates.
Data Source
AI summary
A method for efficiently supporting deletion in a probabilistic data structure, and related computing or storage system are described. A processor, computing system or storage system constructs a table and a summary table for determining whether there is an entry for a value in the table. The summary table has buckets pointed to by address fields of values. Each bucket has a prefix table, a transit table, signature table and a first indicator. The system tracks deletion and addition of items of the table and summary table through the first indicators.