The research group of Professor Hideo Ohno and Associate Professor Shunsuke Fukami of Tohoku University has demonstrated the sub-nanosecond operation of a nonvolatile magnetic memory device.
Recently, the concept of “Internet of Things” (IoT) – a giant network of connected devices, people and things – has been attracting a great deal of attention. Although its range of application is limited at this stage, it is expected that in the near future, IoT will be widely applied and will play important roles in fields such as security, automated driving, social infrastructure and disability aid.
An integrated circuit, or microcontroller unit, is the brain in the IoT society, where information is acquired, processed, and transmitted. Thus, development of device technologies to make integrated circuits ultralow-power and high-performance, or high-speed, is of great importance for the progress of the IoT society.
In terms of low-power, the use of nonvolatile memories is known to be effective.
On the other hand, in terms of high-performance, it has been difficult for the nonvolatile memories which are both currently available (commercialized) and under development (not commercialized yet) to achieve the speed comparable to the one realized with currently-used volatile static random access memories.
The research group at Tohoku University had previously announced that they had developed a new-structure nonvolatile magnetic memory device. The device has a three-terminal configuration, which is different from the two-terminal magnetic memory device that is just about to hit the market.
The device uses a new scheme of spin-orbit torque-induced magnetization switching, which has been predicted to be suitable for the fast control of magnetization.
Here, the group fabricated the developed new-structure device and successfully demonstrated 0.5-nanosecond operation with a sufficiently small current. The achieved speed is comparable to the highest class of static random access memories currently available.
The group members showed that the current required to switch the magnetization does not significantly change with the operation speed unlike the case for the conventional two-terminal magnetic memory devices where the required current increases as the speed increases.
They also addressed several issues of the spin-orbit torque-induced switching device. They achieved an external-field-free switching and a reduction of switching current density by improving the structure and material systems.
The present work is expected to pave the way for the realization of ultralow-power and yet high-performance microcontroller units that are indispensable for the future progress of IoT societies.
The Latest on: Nonvolatile memory
via Google News
The Latest on: Nonvolatile memory
- STMicroelectronics Introduces Safe, Real-Time Microcontrollers for Next-Generation Automotive Domain Architectures on February 21, 2019 at 9:43 pm
The first Stellar automotive MCU samples combine six Arm Cortex-R52 cores with embedded non-volatile memory for fast and deterministic real-time computing, free from the delays of non-integrated ... […]
- Next Generation Memory Market is Expected to Touch a Value of US$ 8.5 Billion by 2024 on February 21, 2019 at 10:42 am
Feb 21, 2019 (Heraldkeeper via COMTEX) -- New York, February 21, 2019: The Next-Generation Memory Market is segmented on the lines of its application, product and regional. Basis of product is segment... […]
- Development of nonvolatile spintronics-based 50uW microcontroller unit operating at 200MHz on February 21, 2019 at 10:12 am
all modules are made non-volatile using spintronics devices, and wasteful power consumption is entirely eliminated by controlling the power supply for each module independently. In addition, the data ... […]
- CEA-Leti & Stanford Target Edge-AI Apps with Breakthrough Memory Cell on February 21, 2019 at 7:02 am
2019 – Researchers at CEA-Leti and Stanford University have developed the world’s first circuit integrating multiple-bit non-volatile memory (NVM) technology called Resistive RAM (RRAM) with silicon c... […]
- Intel confirms non-volatile MRAM is being produced with high yield on February 21, 2019 at 5:36 am
Unlike DRAM used in traditional computer memory, MRAM is non-volatile and can store data for up to 10 years without any power available. That time frame can be extended by storing chips in a cool temp... […]
- Memory Integrated Circuits (IC) Market Revenue and Value Chain 2014-2020 : FMI Study on February 20, 2019 at 5:09 pm
There are primarily two types of ICs for memory storage viz. volatile and non-volatile memory. Volatile memories are memories such as random access memories (RAM) which can retain data stored on them ... […]
- Guide to Flash Storage Memory on February 20, 2019 at 12:16 pm
Let’s define some important terms early on: flash and SSD are closely related, but are not synonymous. Flash: Flash memory is non-volatile computer memory for persistent storage and data transfers bet... […]
- Video: In-Memory Computing Using Photonic Memory Devices on February 20, 2019 at 9:24 am
In this video, IBM researchers describe an all-optical approach to developing direct in-memory multiplication on an integrated photonic device based on non-volatile multilevel phase-change memories. U... […]
- Emerging Memories Team Up With AI on February 20, 2019 at 6:30 am
It is seeking additional members. At the IEEE ISSCC Conference in San Francisco this week researchers at CEA-LETI and Stanford University said that they had developed a circuit that integrated RRAM no... […]
- Unifying On Premises And Clouds With Flash And NVM-Express on February 19, 2019 at 2:51 pm
Several years after first coming onto the scene, NVM-Express continues to get attention from vendors and organizations alike for the promise it holds to ramp up performance and cut into the latency be... […]
via Bing News