• Navigation überspringen
  • Zur Navigation
  • Zum Seitenende
Organisationsmenü öffnen Organisationsmenü schließen
Friedrich-Alexander-Universität Chair of Computer Science 3 CS3
  • FAUZur zentralen FAU Website
  1. Friedrich-Alexander-Universität
  2. Technische Fakultät
  3. Department Informatik
Suche öffnen
    1. Friedrich-Alexander-Universität
    2. Technische Fakultät
    3. Department Informatik
    Friedrich-Alexander-Universität Chair of Computer Science 3 CS3
    Menu Menu schließen
    • Research
    • Teaching
    • Staff
    • Jobs
    1. Startseite
    2. Lectures / Vorlesungen
    3. Resistive RAM and In-Memory Computing

    Resistive RAM and In-Memory Computing

    Bereichsnavigation: Lectures / Vorlesungen
    • Resistive RAM and In-Memory Computing
    • Supercomputing Praktikum

    Resistive RAM and In-Memory Computing

    John Reuben, Ph.D.

    John Reuben Prabahar, Ph.D.

    Post-doctoral researcher

    Department Informatik (INF)
    Lehrstuhl für Informatik 3 (Rechnerarchitektur)

    Raum: Raum 07.139
    Martensstrasse 3
    Erlangen
    Germany
    • Telefon: +49 9131 85-27999
    • E-Mail: johnreuben.prabahar@fau.de
    • Webseite: http://www3.informatik.uni-erlangen.de

    Link to Campo:

    https://www.campo.fau.de:443/qisserver/pages/startFlow.xhtml?_flowId=detailView-flow&unitId=121793&periodId=565&navigationPosition=hisinoneLehrorganisation,examEventOverviewOwn

    Lecturer: Dr.John Reuben Prabahar

    Summary

    Flash memory has been the de-facto standard for NVM (Non-Volatile Memories) for the last two decades and has been used in our computers, USB sticks, mobile phones and almost all electronic devices which need to retain some data without being powered ON. Due to technical limitations faced by flash memory, academia and industry have been researching new NVMs as replacement. Resistive Random Access Memory is one such promising NVM among other emerging Non-Volatile Memories (NVM). This course is aimed as an introductory course on ReRAM and In-Memory computing. First, the students will be introduced to this memory device and its characteristics. Then, memory array design is introduced. The peripheral circuitry i.e. the CMOS circuits around the memory for reading and writing into the memory array are elaborated. After the students become familiarized with the memory technology, they will be introduced to in-memory computing and how certain computing tasks can be implemented in the ReRAM array. In the accompanying exercises, the students will be taught how to simulate the memory array (using a ReRAM device model) and design the peripheral circuits using a transistor-level circuit-simulation tool (Cadence Virtuoso).
    Note:
    1. Students who register for this course must also register for the execise component „ReRAM and In-Memory Computing Exercise“ which is 2.5 ECTS.
    2. It is NOT possible to register for the lecture component alone (2.5 ECTS) or exercise component alone (2.5 ECTS). Both must be registered (5 ECTS together) since the exercises closely follow the lecture.

    Contents:

    1. Concept of memory hierarchy, Advantages, evolution of memory hierarchy to include Storage class memory
    (SCM).
    2. SRAM, the standard for cache memory (Single bit SRAM Cell, READ/WRITE/HOLD operations)
    3. DRAM, the standard for main memory (Single bit DRAM Cell, READ/WRITE/REFRESH operations)
    4. Floating gate transistor (flash), the present standard for tertiary memory (cell structure, Array, READ/WRITE)
    5. ReRAM as the emerging NVM, ReRAM Device Switching Charactersitics, Endurance, Retention,
    charactersitics
    6. Variability and Modeling of ReRAM
    7. Resistive RAM Array Configurations (1R, 1S-1R, 1T-1R), Selectors for 1S–1R Array
    8. Row Decoder, Column Multiplexer for Memory array
    9. Three Ways to Increase Memory Density (Scaling, Multi level Cell, 3D stacking) Writing Multiple states and Reading multiple states
    10. Voltage Mode Sensing, Time-based Sensing
    11. Current-mode Sensing, ReRAM technology-specific issues to be considered while sensing ReRAM
    12. Basic WRITE circuit and Voltage Regulation during WRITE operation,Incremental Program and Verify Approach
    13. WRITE Termination Approach, Writing into 1S-1R arrays, Issues faced while writing into ReRAM cell
    14. Von Neumann Bottleneck (The Memory Wall), In-memory Arithmetic (NOR and MAJORITY gate)
    15. In-Memory Adders
    16. Matrix Vector Multiplication(MVM) in ReRAM Array
    17. Improved MVM using bit slicing

    Recommended Requirements:

    Students are recommended to have a basic knowledge of CMOS
    (Complementary Metal Oxide Semiconductor) transistors, analog electronics
    (differential amplifiers) and digital electronics (Flip-flops, Boolean logic gates).
    This is usually acquired through a basic course on Integrated circuits/ Electronic
    circuit design/ VLSI in the Bachelors/undergraduate level in most universities.

    List degree programmes for which the module is intended:
    – M.Sc. Information and Communication Technology
    – M. Sc. Elektrotechnik, Elektronik und Informationstechnik
    – M. Sc. Informatik
    -M. Sc. Computational Engineering

     

     

    Friedrich-Alexander-Universität
    Lehrstuhl für Informatik 3

    Martensstrasse 3
    91058 Erlangen
    • Impressum
    • Datenschutz
    • Barrierefreiheit
    • RSS Feed
    Nach oben