Networked Power-Gated MRAMs for Memory-Based Computing

Emerging nonvolatile memory technologies open new perspectives for original computing architectures. In this paper, we propose a new type of flexible and energy-efficient architecture that relies on power-gated distributed magnetoresistive random access memory (MRAM). The proposed architecture uses a network-on-chip (NoC) to interconnect MRAM-based clusters, processing elements, and managers. The NoC distributes application-specific commands to MRAM devices by means of packets. Configurable network interfaces allow to transform MRAM devices into smart units able to respond to incoming commands. In this context, three types of MRAM designs are proposed with different power-gating policies and granularities. A relevant database search engine case study is considered to illustrate the benefits of this proposed architecture. It is implemented with a sparse-neural-network approach and simulated in SystemC with different scenarios including hundreds of database queries. Hardware designs and accurate power estimations have been conducted. The obtained results demonstrate important power reduction with database hit rates of about 94%. Targeting 65-nm technology, energy savings reach 87% when compared with an static random access memory-based implementation. Moreover, a new asymmetric read/write MRAM type provides from 39% to 50% energy reduction with respect to the other fixed-granularity models. This results in a low-power, highly scalable, and configurable implementation of memory-based computing.

[1]  Daisuke Suzuki,et al.  Standby-Power-Free Integrated Circuits Using MTJ-Based VLSI Computing , 2016, Proceedings of the IEEE.

[2]  Vincent Gripon,et al.  Sparse Neural Networks With Large Learning Diversity , 2011, IEEE Transactions on Neural Networks.

[3]  Swarup Bhunia,et al.  Memory-based computing for performance and energy improvement in multicore architectures , 2012, GLSVLSI '12.

[4]  Kiyoung Choi,et al.  Active Memory Processor for Network-on-Chip-Based Architecture , 2012, IEEE Transactions on Computers.

[5]  Luca Benini,et al.  Energy-efficient GPGPU architectures via collaborative compilation and memristive memory-based computing , 2014, 2014 51st ACM/EDAC/IEEE Design Automation Conference (DAC).

[6]  Martha Johanna Sepúlveda,et al.  Scalable NoC-based architecture of neural coding for new efficient associative memories , 2013, 2013 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS).

[7]  Hui Zhao,et al.  A Scaling Roadmap and Performance Evaluation of In-Plane and Perpendicular MTJ Based STT-MRAMs for High-Density Cache Memory , 2013, IEEE Journal of Solid-State Circuits.

[8]  Hiroki Koike,et al.  High-speed simulator including accurate MTJ models for spintronics integrated circuit design , 2012, 2012 IEEE International Symposium on Circuits and Systems.

[9]  Chein-Wei Jen,et al.  On the design automation of the memory-based VLSI architectures for FIR filters , 1993 .

[10]  S. Bhunia,et al.  A Scalable Memory-Based Reconfigurable Computing Framework for Nanoscale Crossbar , 2012, IEEE Transactions on Nanotechnology.

[11]  Vincent Gripon,et al.  NoC-MRAM architecture for memory-based computing: Database-search case study , 2017, 2017 15th IEEE International New Circuits and Systems Conference (NEWCAS).

[12]  H. Ohno,et al.  A 1 Mb Nonvolatile Embedded Memory Using 4T2MTJ Cell With 32 b Fine-Grained Power Gating Scheme , 2013, IEEE Journal of Solid-State Circuits.

[13]  E. Culurciello,et al.  NeuFlow: Dataflow vision processing system-on-a-chip , 2012, 2012 IEEE 55th International Midwest Symposium on Circuits and Systems (MWSCAS).

[14]  Martha Johanna Sepúlveda,et al.  Notifying memories: A case-study on data-flow applications with NoC interfaces implementation , 2016, 2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC).

[15]  Jason Cong,et al.  Invited: Heterogeneous datacenters: Options and opportunities , 2016, 2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC).

[16]  H. Ohno,et al.  A perpendicular-anisotropy CoFeB-MgO magnetic tunnel junction. , 2010, Nature materials.

[17]  Shoji Ikeda,et al.  A 32-Mb SPRAM With 2T1R Memory Cell, Localized Bi-Directional Write Driver and `1'/`0' Dual-Array Equalized Reference Scheme , 2010, IEEE Journal of Solid-State Circuits.

[18]  Giovanni De Micheli,et al.  The Programmable Logic-in-Memory (PLiM) computer , 2016, 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[19]  H. Ohno,et al.  Fabrication of a Nonvolatile Full Adder Based on Logic-in-Memory Architecture Using Magnetic Tunnel Junctions , 2008 .

[20]  Shaahin Angizi,et al.  In-Memory Computing with Spintronic Devices , 2017, 2017 IEEE Computer Society Annual Symposium on VLSI (ISVLSI).

[21]  Chankyung Kim,et al.  7.4 A covalent-bonded cross-coupled current-mode sense amplifier for STT-MRAM with 1T1MTJ common source-line structure array , 2015, 2015 IEEE International Solid-State Circuits Conference - (ISSCC) Digest of Technical Papers.

[22]  Andrew S. Cassidy,et al.  A million spiking-neuron integrated circuit with a scalable communication network and interface , 2014, Science.

[23]  Vincent Gripon,et al.  A Nonvolatile Associative Memory-Based Context-Driven Search Engine Using 90 nm CMOS/MTJ-Hybrid Logic-in-Memory Architecture , 2014, IEEE Journal on Emerging and Selected Topics in Circuits and Systems.

[24]  G. Huang,et al.  An Energy-Efficient Nonvolatile In-Memory Computing Architecture for Extreme Learning Machine by Domain-Wall Nanowire Devices , 2015, IEEE Transactions on Nanotechnology.

[25]  Kang L. Wang,et al.  Interfacial Perpendicular Magnetic Anisotropy in Sub-20 nm Tunnel Junctions for Large-Capacity Spin-Transfer Torque Magnetic Random-Access Memory , 2017, IEEE Magnetics Letters.

[26]  Shoji Ikeda,et al.  Properties of magnetic tunnel junctions with a MgO/CoFeB/Ta/CoFeB/MgO recording structure down to junction diameter of 11 nm , 2014 .