dorsaVi Launches Strategic Collaboration with ITRI and NTU to Advance 22nm RRAM Technology

dorsaVi Limited (ASX: DVL) has entered into a multi-year collaboration with Taiwan’s Industrial Technology Research Institute (ITRI) and Singapore’s Nanyang Technological University (NTU) to develop its next-generation 22-nanometre Resistive Random Access Memory (RRAM) platform.

This partnership builds on dorsaVi’s existing relationship with NTU and aims to propel the company’s RRAM architecture to the 22-nanometre technology node. ITRI, a globally recognised research institution with over 6,500 employees, brings expertise in circuit design and backend-of-line processing, while NTU contributes its advanced RRAM device and materials research.

The initiative is anchored on NTU’s exclusively licensed RRAM intellectual property and is intended to deliver lower energy usage, faster switching speeds, and higher memory density. The 22nm platform is being developed for full compatibility with TSMC’s 22nm CMOS manufacturing processes, facilitating broad commercial adoption.

dorsaVi is targeting several applications with this technology, including robotics, wearables, electric vehicles, industrial automation, neuromorphic computing, and ultra edge artificial intelligence. The architecture combines in-memory computing and neuromorphic features, aiming to enable real-time, low-power data processing for embedded systems.

Robotics has been identified as the lead commercial opportunity, with projected growth in the global market expected to reach USD 372.59 billion by 2034. The company’s platform is designed to support motor control, in-sensor processing, and always-on systems for intelligent automation.

Performance metrics for the 22nm RRAM program include a write latency of 100 to 200 nanoseconds, write voltages under 2.0 volts, and endurance exceeding 10 million cycles. Target deliverables include 1Mbit RRAM macros with write-verify capability and multi-state compute-in-memory structures with over 20 TOPS per watt energy efficiency.

The collaboration is expected to run for an initial period of two to three years and marks a key step in dorsaVi’s strategy to build a unified platform for ultra edge intelligence.