Fig. 4

Comparison of Time Consumption Using Bloom Filters (BF), B-tree Indexing and Direct Server Search in a Commercial Database. The analysis for Bloom filters is broken down into individual steps: a) Conversion of SMILES notation to an RDKit Mol object, b) Canonicalization of the string. B-tree index was created with PostgreSQL. The time requirements for each step are displayed in the first and third columns. Bars 3 and 4, showcase how the Direct Server Search time is influenced by network latency, a common concern when relying on remote, online resources