MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202611101526 A) filed by Chandigarh University on August 22, 2026, for Integrated Processing-In-Memory System For Graph Execution And Vector Similarity Search.
Inventors include Shaman Sharma; Harbinder Singh; Nikita Singh; and Deepak Kumar.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: The present invention discloses an integrated processing-in-memory (PIM) system for unified graph execution and high-dimensional vector similarity search. The system comprises a non-volatile resistive random-access memory (ReRAM) crossbar array configured to store graph and vector data and perform mixed-signal vector- matrix multiplication within the memory array. A hardware controller comprises a dynamic graph execution unit and a high-dimensional vector search engine for performing graph traversal and vector similarity computations. An on-chip spiking neural network (SNN) accelerator having spike-timing-dependent plasticity (STDP) circuitry monitors memory access characteristics and controls dynamic data remapping within the ReRAM array. A database software interface compiles graph and vector queries into hardware instructions for direct execution by the PIM architecture, thereby reducing off-chip data movement and enabling coordinated graph and vector processing.
Disclaimer: Curated by HT Syndication.
