Recent MRAM News - Page 2

Researchers from  Purdue University were awarded with a new project to advanced AI using research performed at Purdue for over a decade. The CHEETA:CMOS+MRAM project will utilize a CMOS+X approach specifically leveraging the unique capabilities of MRAM memory to design efficient in-memory computing hardware fabrics.

The CHEETA project aims to deliver impactful improvements in energy efficiency and sensor-to-decision latency compared to the current commercial standard. The project received funding from the Applied Research Institute. 

Researchers from the Institute of Science Tokyo (formerly Tokyo Tech), in collaboration with Western Digital, has developed a new high-performance material suitable for low-power SOT-MRAM.

The researchers are using BiSb material under the CoFeB/MgO, with perpendicular magnetic anisotropy. By optimizing the interfacial layer thickness as well as deposition condition of BiSb, the researchers managed to achieve a large effective spin Hall angle and relatively high electrical conductivity. The researchers continued to demonstrate spin–orbit torque-induced magnetization switching driven by a small threshold current density.

Everspin Technologies announced that Lucid Motors has chosen the company's PERSYST MRAM technology to be used in its latest all-electric Gravity SUV. 

Lucid's Gravity uses Everspin's MR25H256A, a 256Kb serial MRAM. Everspin says it was selected because it meets the AEC Q100 Grade 1 specification of -40C to 125C temperature operation. The MRAM is used in the Gravity to handle data logging and parameter storage to assist in the efficient operation of the all-electric powertrain.

Everspin Technologies is at the forefront of the MRAM industry, with a product range that spans Toggle MRAM and STT-MRAM, embedded MRAM IP, more. The company's president and CEO, Sanjeev Aggarwal, was kind enough to answer a few questions we had about the company, it's recent announcements and its future roadmap.

Thank you for your time Sanjeev, we appreciate it. A few days ago, Everspin announced it had received a strategic $14.55 million 2.5-year DoD project to provide continued and stable manufacturing services. Can you share more info? How will that money be allocated, and how will you ensure stable support?

Everspin has partnered with the DoD on various projects to deploy our MRAM technology and fab MRAM products for Defense Industrial Base customers in Chandler, Arizona. The award money will be used to strength the US MRAM manufacturing supply chain.  Everspin will look at its supply chain and mitigate any risks identified. This can include second-sourcing gases and chemicals as well as upgrading processes.  

US-based Carnegie Mellon University spinoff Efficient has signed a strategic partnership with GlobalFoundries to launch its low-power Fabric chip on GlobalFoundries' 22FDX low power silicon on insulator process. Efficient will adopt embedded MRAM in its chip, and expects to launch it (first samples) by the summer of 2025.

Efficient is an early stage company that raised over $16 million to develop the power-efficient Fabric processor, targeting edge applications such as machine learning-enabled extreme-edge machine vision, continuous audio intelligence and versatile sensory and signals intelligence.

GlobalFoundries has been offering eMRAM 22nm process solutions since 2017, based on Everspin's pMTJ technology and IP.

Everspin Technologies announced that it has received a strategic $14.55 million 2.5-years project from the Department of Defense (DoD) to provide continued and stable manufacturing services for aerospace and defense segments. Under the award, Everspin will provide a plan to mitigate risks to its MRAM manufacturing supply chain.

Everspin is producing its MRAM products at its manufacturing line in Chandler, Arizon. In March 2024, Everspin announced that it has applied for the CHIPS Incentives Program, asking for funding for an additional 200mm MRAM production line. It seems as if this current DoD award is not that project that was applied to in March. In 2022, Everspin asked for funding to build a new MRAM production line in Indiana.

Everspin Technologies reported its financial results for Q2 2022. Revenues were $10.6 million (down from $15.7 million in Q2 2023), and net loss was $0.2 million (down from a new profit of $5.4 million in Q2 2023). The company is cash flow positive, generated $1.7 million in the quarter, and ended it with $36.8 million in cash and equivalents.

During the quarter, the company experienced weakness in toggle MRAM sales, partly due to inventory consumption and partly due to unfavorable currency exchange. Looking ahead, Everspin sees positive signs of inventory consumption, and expects a modest ramp in toggle MRAM sales in the rest of 2024. Everspin said it has strong traction with its new PERSYST family of MRAM products, and expects to start delivering the first products to its customers towards the end of 2024 or early 2025.

pMTJ STT-MRAM developer Avalanche Technology launched a new enhanced discrete product family called "Space Grade-E", targeting the aerospace & defense community. The new Space Grade-E MRAM is based on Avalancehe's Gen 3 high density, high reliability MRAM technology, and it allows pin compatible scalability to MEO, GEO, and beyond.

The Space Grade-E family is now sampling, and is offered in both Parallel and Dual QSPI interface types. Compared to the company's previous generation space-grade MRAM, this new generation includes a boost in radiation tolerance specs and burn in hours for improved reliability. The new memory is maintains pin-compatible migration from Avalanche's previous generation space-grade MRAM. 

Researchers from the University of Minnesota Twin Cities developed a new AI hardware accelerator, based on MRAM memory, that can accelerate AI computation by a factor of 1,000. 

The so-calledd Computational Random-Access Memory device, or CRAM, is a machine learning inference accelerator, that is not just 1,000 faster, it also exhibits energy saves of 1,700 to 2,500 times compared to traditional methods.