In-Memory Associative Processors: Tutorial, Potential, and Challenges

被引:7
作者
Fouda, Mohammed E. [1 ,2 ]
Yantir, Hasan Erdem [3 ]
Eltawil, Ahmed M. [4 ]
Kurdahi, Fadi [1 ]
机构
[1] Univ Calif Irvine, Ctr Embedded & Cyber Phys Syst, Irvine, CA 92697 USA
[2] Nile Univ, Nanoelect Integrated Syst Ctr, Giza 16453, Egypt
[3] TUBITAK Informat & Informat Secur Res Ctr, TR-41470 Kocaeli, Turkey
[4] King Abdullah Univ Sci & Technol, Elect & Comp Engn Dept, Thuwal 23955, Saudi Arabia
关键词
Program processors; Computer architecture; Table lookup; Transistors; Switches; Performance evaluation; Associative memory; Associative processor; in-memory computing; content-addressable memory; accelerators; SIMD; vector processor; ARCHITECTURES; INFERENCE;
D O I
10.1109/TCSII.2022.3170468
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
In-memory computing is an emerging computing paradigm that overcomes the limitations of exiting Von-Neumann computing architectures such as the memory-wall bottleneck. In such paradigm, the computations are performed directly on the data stored in the memory, which highly reduces the memory-processor communications during computation. Hence, significant speedup and energy savings could be achieved especially with data-intensive applications. Associative processors (APs) were proposed in the seventies and recently were revived thanks to the high-density memories. In this tutorial brief, we overview the functionalities and recent trends of APs in addition to the implementation of each content-addressable memory with different technologies. The AP operations and runtime complexity are also summarized. We also explain and explore the possible applications that can benefit from APs. Finally, the AP limitations, challenges, and future directions are discussed.
引用
收藏
页码:2641 / 2647
页数:7
相关论文
共 55 条
  • [31] Content-addressable memory (CAM) circuits and architectures: A tutorial and survey
    Pagiamtzis, K
    Sheikholeslami, A
    [J]. IEEE JOURNAL OF SOLID-STATE CIRCUITS, 2006, 41 (03) : 712 - 727
  • [32] Tree-based machine learning performed in-memory with memristive analog CAM
    Pedretti, Giacomo
    Graves, Catherine E.
    Serebryakov, Sergey
    Mao, Ruibin
    Sheng, Xia
    Foltin, Martin
    Li, Can
    Strachan, John Paul
    [J]. NATURE COMMUNICATIONS, 2021, 12 (01)
  • [33] Design Exploration of Sensing Techniques in 2T-2R Resistive Ternary CAMs
    Rakka, M.
    Fouda, M. E.
    Kanj, R.
    Eltawil, A.
    Kurdahi, F. J.
    [J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2021, 68 (02) : 762 - 766
  • [34] Sasan A, 2009, DES AUT TEST EUROPE, P911
  • [35] A RECONFIGURABLE FULLY PARALLEL ASSOCIATIVE PROCESSOR
    SCHERSON, ID
    ILGEN, S
    [J]. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1989, 6 (01) : 69 - 89
  • [36] SCHUSTER SA, 1979, IEEE T COMPUT, V28, P446, DOI 10.1109/TC.1979.1675383
  • [37] ISAAC: A Convolutional Neural Network Accelerator with In-Situ Analog Arithmetic in Crossbars
    Shafiee, Ali
    Nag, Anirban
    Muralimanohar, Naveen
    Balasubramonian, Rajeev
    Strachan, John Paul
    Hu, Miao
    Williams, R. Stanley
    Srikumar, Vivek
    [J]. 2016 ACM/IEEE 43RD ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE (ISCA), 2016, : 14 - 26
  • [38] Smagulova K., 2021, ARXIV210903934
  • [39] STORMON CD, 1992, IEEE MICRO, V12, P68, DOI 10.1109/40.180249
  • [40] Thurber K. J., 1975, Computing Surveys, V7, P215, DOI 10.1145/356654.356659