W02 Computation-In-Memory (CIM): from Device to Applications
Aim of the Conference: All issues with which the architectures and technologies are face today have led to the slowdown of the traditional device scaling. In order for computing systems to continue deliver sustainable benefits for the foreseeable future, alternative computing architectures and notions have to be explored in the light of emerging new device technologies. This workshop aims at providing a forum to discuss Computation-in-Memory (as an alternative architecture) in the light of emerging non-volatile devices (such as RRAM, PCM and STT-MRAM), and its potential applications. It also aims at reinforcing the CIM community and at offering a holistic vision of this emerging computing paradigm to the electronic design, automation and test communities.
The authors of accepted papers will be invited to submit an extended version of their work to the Special Issue of ACM Journal of Emerging Technologies in Computing.
The workshop covers all aspects of CIM based on non-volatile devices including (but not limited to):
- Device and technology: physics and modeling, device technologies, device characterization.
- Novel logic and circuit design concepts using NV devices: Boolean logic, threshold logic, arithmetic circuits, multi-level based logic, memories, PUF technology, TRNG design.
- System architectures and new computing paradigms: resistive computing, neuro-inspired computing, novel architectures and CMOS integration, cellular automata and array computing.
- Applications exploiting NV devices: signal processing, chaos and complex networks, sensors applications, AI applications.
- Automation and CAD tools: mapping tools, compilers, logic synthesis tools, design space exploration tools.
- Test and Reliability: test and reliability solutions for circuits and architectures.
The workshop will be co-located with the DATE 2021 conference. At least one author of each accepted paper is expected to register to the workshop using DATE’s regular registration system.
Opening Session Keynote Address
Alberto Bosio (INL, France), Said Hamdioui (TUDelft, The Netherlands), Elena-Ioana Vatajelu (TIMA, France)
To be announced
Invited Talk Invited Talk
Panel Computation-In-Memory (CIM): from Device to Applications
Panelists (preliminary list):
- Henri-Pierre Charles (CEA-Leti, FR)
- Shahar Kvatinsky (Technion, Israel)
- Alexandre Levisse (EPFL, CH)
- Georgios Sirakoulis (University of Thrace, GR)
Session Scientific Presentations
Impact of On-Chip Interconnect on In-Memory Acceleration of Deep Neural Networks
Gokul Krishnan*, Sumit K. Mandal*, Chaitali Chakrabarti, Jae-sun Seo, Umit Y. Ogras, Yu Cao (School of Electrical, Computer and Energy Engineering, Arizona State University)
Conversion-in-Memory Using Floating-Gate Memristive Neural Networks
Loai Danial, Shahar Kvatinsky (Viterbi Faculty of Electrical Engineering, Technion - Israel Institute of Technology)
Exact Stochastic Computing Multiplication in Memristive Memory
Mohsen Riahi Alam, M. Hassan Najafi (University of Louisiana at Lafayette, Louisiana, USA), Nima TaheriNejad (TU Wien, Vienna, Austria)
Efficient 8-bit matrix multiplication on Computational SRAM architecture
Mambu Kévin, Charles Henri-Pierre, Kooli Maha (Université Grenoble Alpes, CEA LIST, F-38000 Grenoble, France)
Session Scientific Presentations
Simulation and experimental characterization of memristive crossbar arrays for computation-in-memory
J. Mohr, C. Bengel, M. Abu, S.Hamdioui, D.J. Wouters, S. Menzel, R. Waser
Sensing and Processing in Memristive Arrays
Saurabh Khandelwal (School of ECM, Oxford Brookes University, Oxford), Shahar Kvatinsky (Viterbi Faculty of Electrical Engg., Technion - Israel Institute of Technology), Marco Ottavi, Eugenio Martinelli (Electronic Engg., University of Rome “Tor Vergata”, Rome, Italy), Abusaleh Jabir (School of ECM, Oxford Brookes University, Oxford)
Exploration of a Scalable Vector-Tile-based In-Memory Computing Architecture
Roman Gauchi, Valentin Egloff, Maha Kooli, Jean-Philippe Noel, Bastien Giraud, Pascal Vivet (Université Grenoble Alpes, CEA LIST, Grenoble, France), Subhasish Mitra (Stanford University, Palo Alto, CA, USA), Henri-Pierre Charles (Université Grenoble Alpes, CEA LIST, Grenoble, France)
Reuse-aware architecture and synthesis flow for NVM-based FPGAs
João Paulo Cardoso de Lima, Rafael Fão de Moura, Luigi Carro (Departamento de Informática Aplicada Instituto de Informática - UFRGS)
Toward CIM architecture with a new tile-level simulator
Mahdi Zahedi, Stephan Wong, Said Hamdioui (Computer Engineering Laboratory, Delft University of Technology, The Netherlands)