About this Research Topic
Semiconductor memory is one of the footstones of modern computing systems. Due to the rapidly expanding needs for computing and storage, system scaling and performance enhancement enabled by merely Moore's law are facing significant challenges. The industry and research community have been actively exploring alternative memory technologies, such as magnetic memory (MRAM), resistive memory (RRAM), ferroelectric memory (FeRAM/FeFET), phase change memory (PCRAM), as well as novel memory devices based on oxide semiconductors and 2D materials. The ultimate goal is not only to pursue the “ideal” memory technology that features improved performance and scalability at reduced energy consumption and manufacture-cost, but also to overcome the bottleneck of “memory wall” for applications such as artificial intelligence (AI) and Internet-of-things (IoT), where high performance, high density, and high energy efficiency are desperately needed.
For alternative memories aiming at replacing existing solutions, they should provide comparable performance with the existing memory technologies at overwhelmingly higher density, or offer overwhelmingly better performance at comparable density. There comes the eternal demand for memory, that is higher density at reduced cost per bit. Moreover, the data-intensive applications such as machine learning also require higher access speed at lower energy consumption and even collocated memory and computing. However, the recent speaking of “universal memory” to replace SRAM, DRAM, Flash, and other conventional memories is barely possible in practice. Therefore, we need to review the true advantage of emerging memories and their applications in new computing paradigms, making efforts to foster innovations and solutions not only from levels of material, device, and technology, but also levels of circuit, architecture, and algorithm, in order to revolutionize memory technologies as well as the memory-centric computing architecture.
This general Research Topic aims to bring together state-of-the-art interdisciplinary researches in the above context including enabling technologies for the plethora of alternative memories and new computing paradigms such as processing-in-memory (PIM) and neuromorphic computing. We invite original research, review, mini-review, and perspective articles on topics that include, but are not limited to the following research interests:
- New materials for emerging memory technology, such as 2D materials, oxide semiconductors, etc.
- Emerging non-volatile memories, such as MRAM, PCRAM, ReRAM, FeRAM/FeFET, etc.
- Novel volatile memory concepts, such as 2T0C DRAM, vertical DRAM, etc.
- Advanced integration and packaging technologies for memory, such as 2.5D/3D integration, chiplet, etc.
- New computing paradigms based on memory devices and systems for AI, such as processing-in-memory, logic-in-memory, neuromorphic computing
- Issues and countermeasures for reliability, energy efficiency, fault tolerance, etc.
- Modeling and simulation of memory devices
- EDA design and Simulation tools/platforms for memory-based computing paradigm
- Design-technology co-optimization (DTCO) and system-technology co-optimization (STCO) for high-performance memory and computing systems
For alternative memories aiming at replacing existing solutions, they should provide comparable performance with the existing memory technologies at overwhelmingly higher density, or offer overwhelmingly better performance at comparable density. There comes the eternal demand for memory, that is higher density at reduced cost per bit. Moreover, the data-intensive applications such as machine learning also require higher access speed at lower energy consumption and even collocated memory and computing. However, the recent speaking of “universal memory” to replace SRAM, DRAM, Flash, and other conventional memories is barely possible in practice. Therefore, we need to review the true advantage of emerging memories and their applications in new computing paradigms, making efforts to foster innovations and solutions not only from levels of material, device, and technology, but also levels of circuit, architecture, and algorithm, in order to revolutionize memory technologies as well as the memory-centric computing architecture.
This general Research Topic aims to bring together state-of-the-art interdisciplinary researches in the above context including enabling technologies for the plethora of alternative memories and new computing paradigms such as processing-in-memory (PIM) and neuromorphic computing. We invite original research, review, mini-review, and perspective articles on topics that include, but are not limited to the following research interests:
- New materials for emerging memory technology, such as 2D materials, oxide semiconductors, etc.
- Emerging non-volatile memories, such as MRAM, PCRAM, ReRAM, FeRAM/FeFET, etc.
- Novel volatile memory concepts, such as 2T0C DRAM, vertical DRAM, etc.
- Advanced integration and packaging technologies for memory, such as 2.5D/3D integration, chiplet, etc.
- New computing paradigms based on memory devices and systems for AI, such as processing-in-memory, logic-in-memory, neuromorphic computing
- Issues and countermeasures for reliability, energy efficiency, fault tolerance, etc.
- Modeling and simulation of memory devices
- EDA design and Simulation tools/platforms for memory-based computing paradigm
- Design-technology co-optimization (DTCO) and system-technology co-optimization (STCO) for high-performance memory and computing systems
Keywords: Memory technology, Processing-in-memory (PIM), neuromorphic computing, MRAM, PCRAM, ReRAM, FeRAM/FeFET, DRAM
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
