Salted Hybrid Bloom Filter (SHBF)

Kapil N, Thakral C, Raghuvanshi A and Sathiyamoorthy E

Published on: 2025-08-20

Abstract

Big Data’s growth led to the change of technological environment in the fields of IT, Bioinformatics etc. There is an apparent spike in the solution requirement for efficient data management in the era of continuous data generation. Bloom Filter is a probabilistic data structure that efficiently handles big datasets, which filters out redundant data and makes effective use of memory. One of the major drawbacks of Bloom Filtering is False Positives, which gives rise to operational inefficiencies and secondary lookups.

Types of Bloom Filter techniques, including Cuckoo Filter, Partitioned Bloom Filter, and Counting Bloom filter have been recommended as solution. Main goals of these techniques include optimizing the spatial properties and increasing the number of bits allotted to virtual Bloom array. In this work we have defined how the Modular Dual Bloom Filter architecture works. It involves dual fingerprint verification along with the salting property that helps in significantly reducing the false positives which requires two independent fingerprint checks for confirmation. This algorithm uses modular partitioning with multilevel fingerprint arrays instead of only bit array.