Architectures and Design Methodologies for Micro and Nanocomputing



Similar documents
Agenda. Michele Taliercio, Il circuito Integrato, Novembre 2001

An Open Architecture through Nanocomputing

Architectural Level Power Consumption of Network on Chip. Presenter: YUAN Zheng

Nanocomputer & Architecture

Example-driven Interconnect Synthesis for Heterogeneous Coarse-Grain Reconfigurable Logic

Gates, Circuits, and Boolean Algebra

McPAT: An Integrated Power, Area, and Timing Modeling Framework for Multicore and Manycore Architectures

Chapter 4 Register Transfer and Microoperations. Section 4.1 Register Transfer Language

Two-Phase Clocking Scheme for Low-Power and High- Speed VLSI

Power Reduction Techniques in the SoC Clock Network. Clock Power

Implementation of Full -Parallelism AES Encryption and Decryption

Exploiting Stateful Inspection of Network Security in Reconfigurable Hardware

Aims and Objectives. E 3.05 Digital System Design. Course Syllabus. Course Syllabus (1) Programmable Logic

Introduction to Digital System Design

Outline. Introduction. Multiprocessor Systems on Chip. A MPSoC Example: Nexperia DVP. A New Paradigm: Network on Chip

International Journal of Electronics and Computer Science Engineering 1482

How To Design A Single Chip System Bus (Amba) For A Single Threaded Microprocessor (Mma) (I386) (Mmb) (Microprocessor) (Ai) (Bower) (Dmi) (Dual

Photonic components for signal routing in optical networks on chip

PROGETTO DI SISTEMI ELETTRONICI DIGITALI. Digital Systems Design. Digital Circuits Advanced Topics

A Lab Course on Computer Architecture

Assessment Plan for CS and CIS Degree Programs Computer Science Dept. Texas A&M University - Commerce

Implementation Details

A bachelor of science degree in electrical engineering with a cumulative undergraduate GPA of at least 3.0 on a 4.0 scale

Testing of Digital System-on- Chip (SoC)

ASYNCHRONOUS COUNTERS

Topics of Chapter 5 Sequential Machines. Memory elements. Memory element terminology. Clock terminology

List of courses MEngg (Computer Systems)

Chapter 07: Instruction Level Parallelism VLIW, Vector, Array and Multithreaded Processors. Lesson 05: Array Processors

Design and Implementation of an On-Chip timing based Permutation Network for Multiprocessor system on Chip

INTRODUCTION TO DIGITAL SYSTEMS. IMPLEMENTATION: MODULES (ICs) AND NETWORKS IMPLEMENTATION OF ALGORITHMS IN HARDWARE

7a. System-on-chip design and prototyping platforms

SRAM Scaling Limit: Its Circuit & Architecture Solutions

Multistage Interconnection Network for MPSoC: Performances study and prototyping on FPGA

A 1-GSPS CMOS Flash A/D Converter for System-on-Chip Applications

How To Design A Chip Layout

Scalability and Classifications

數 位 積 體 電 路 Digital Integrated Circuits

System on Chip Design. Michael Nydegger

Best Practises for LabVIEW FPGA Design Flow. uk.ni.com ireland.ni.com

BINARY CODED DECIMAL: B.C.D.

From Bus and Crossbar to Network-On-Chip. Arteris S.A.

1. True or False? A voltage level in the range 0 to 2 volts is interpreted as a binary 1.

AMAGNETIC TUNNEL JUNCTION (MTJ) is a vertical

FPGA. AT6000 FPGAs. Application Note AT6000 FPGAs. 3x3 Convolver with Run-Time Reconfigurable Vector Multiplier in Atmel AT6000 FPGAs.

Implementation and Design of AES S-Box on FPGA

Static-Noise-Margin Analysis of Conventional 6T SRAM Cell at 45nm Technology

Serial port interface for microcontroller embedded into integrated power meter

Switched Interconnect for System-on-a-Chip Designs

Rosaria Rinaldi. Dipartimento di Matematica e Fisica «E. De Giorgi» Università del Salento. Scuola Superiore ISUFI Università del Salento

CSE140 Homework #7 - Solution

Testing & Verification of Digital Circuits ECE/CS 5745/6745. Hardware Verification using Symbolic Computation

Lab 1: Introduction to Xilinx ISE Tutorial

SIM-PL: Software for teaching computer hardware at secondary schools in the Netherlands

Phase Change Memory for Neuromorphic Systems and Applications

QCADesigner: A Rapid Design and Simulation Tool for Quantum-Dot Cellular Automata

Chapter 9 Semiconductor Memories. Jin-Fu Li Department of Electrical Engineering National Central University Jungli, Taiwan

Low Power AMD Athlon 64 and AMD Opteron Processors

Multipliers. Introduction

Chapter 12: Multiprocessor Architectures. Lesson 04: Interconnect Networks

Design and Verification of Nine port Network Router

Architectures and Platforms

Interconnection Networks. Interconnection Networks. Interconnection networks are used everywhere!

路 論 Chapter 15 System-Level Physical Design

Open Flow Controller and Switch Datasheet

Lezione 10 Introduzione a OPNET

Low-Overhead Hard Real-time Aware Interconnect Network Router

Testing Low Power Designs with Power-Aware Test Manage Manufacturing Test Power Issues with DFTMAX and TetraMAX

Leakage Power Reduction Using Sleepy Stack Power Gating Technique

Feed Forward Loops in Biological Systems

PhD Student Marco Maggiali

Chapter 2 Heterogeneous Multicore Architecture

PLAS: Analog memory ASIC Conceptual design & development status

Modeling Sequential Elements with Verilog. Prof. Chien-Nan Liu TEL: ext: Sequential Circuit

X 4 CONFIDENTIAL X 4 OTHER PROGRAMME: CUSTOMER: CONTRACT NO.: WPD NO.: DRD NO.: CONTRACTUAL DOC.:

Assessment for Master s Degree Program Fall Spring 2011 Computer Science Dept. Texas A&M University - Commerce

DEVELOPMENT OF DEVICES AND METHODS FOR PHASE AND AC LINEARITY MEASUREMENTS IN DIGITIZERS

ADVANCED APPLICATIONS OF ELECTRICAL ENGINEERING

Systolic Computing. Fundamentals

Design Verification and Test of Digital VLSI Circuits NPTEL Video Course. Module-VII Lecture-I Introduction to Digital VLSI Testing

Chapter 2 Logic Gates and Introduction to Computer Architecture

7 OUT1 8 OUT2 9 OUT3 10 OUT4 11 OUT5 12 OUT6 13 OUT7 14 OUT8 15 OUT9 16 OUT10 17 OUT11 18 OUT12 19 OUT13 20 OUT14 21 OUT15 22 OUT16 OUT17 23 OUT18

2011, The McGraw-Hill Companies, Inc. Chapter 5

Conditioned Reflex Mimic Circuit Design

An On-chip Security Monitoring Solution For System Clock For Low Cost Devices

Memory Elements. Combinational logic cannot remember

An Extension to DNA Based Fredkin Gate Circuits: Design of Reversible Sequential Circuits using Fredkin Gates

The Designer's Guide to VHDL

Floating Point Fused Add-Subtract and Fused Dot-Product Units

SOC architecture and design

RN-coding of Numbers: New Insights and Some Applications

NAME AND SURNAME. TIME: 1 hour 30 minutes 1/6

A Novel Low Power, High Speed 14 Transistor CMOS Full Adder Cell with 50% Improvement in Threshold Loss Problem

CHARGE pumps are the circuits that used to generate dc

VHDL Test Bench Tutorial

Transcription:

Architectures and Design Methodologies for Micro and Nanocomputing PhD Poster Day, December 4, 2014 Matteo Bollo 1 (ID: 24367, I PhD Year) Tutor: Maurizio Zamboni 1 Collaborators: Mariagrazia Graziano 1, Marco Vacca 1, Giovanna Turvani 1 1 VLSI Group

Attended Classes Class Date CFU Parallel and distributed computing 04/07/2014 5 Tecniche avanzate per il progetto di sistemi 18/07/2014 4 elettronici ad alta affidabilità Microelectronic systems 08/09/2014 6 Software and Routing Technologies in Future 26/09/2014 4 Networks (didattica di eccellenza) Data mining: concetti e algoritmi 07/10/2014 4 Un metodo alternativo per l insegnamento 17/10/2014 2 della fisica Fondi competitivi per la ricerca: dall idea alla 06/11/2014 2 scrittura del progetto Nanocomputing: dispositivi, circuiti e architetture following 8 Architectures for Nanocomputing 2 / 15

Research Context and Motivation Most of data elaboration circuits are implemented over CMOS. The high integration level rises many problems in terms of Power Consumption (Leakage and Switching activity currents). Current flowing from VDD to ground generates Side-Channel signals. Side-Channel contains sensible information on the data elaborated and it is an hard issue for cryptography applications. REDUCING POWER AND INCREASING SIDE-CHANNEL ENTROPY with a change of paradigm in technology: is it a solution? Architectures for Nanocomputing 3 / 15

Research Context and Motivation Magnetic Tunnel Junction Devices are: Created as Memory cell in MRAM and reused as Logic Device. Three possible states 0, 1 and RESET are representable, the state could be forced through: External magnetic field. Forcing current in Bit Line and Source Line Each MTJ could be isolated lowering the corresponding Word Line to 0. The power consumption does not depend on the data computed it is constant. Architectures for Nanocomputing 4 / 15

Magnetic Tunnel Junction use Magnetic Tunnel Junction are very suitable for crossbar (matrix) structure. Architectures for Nanocomputing 5 / 15

Is it possible to make a logic simulation of MTJ? Implementing this FSM model in VHDL we are able to simulate the Magnetic Tunnel Junction logic behaviour. Architectures for Nanocomputing 6 / 15

Addressed research questions/problems Starting from the single device model, is it possible to make a logic simulation of MTJ circuits? Architectures for Nanocomputing 7 / 15

Novel Contributions Development of a design tool and simulation environment with: The capability to integrate a very large number of Nano devices in short time: We have implemented the 4-bit ripple carry adder, the 4-bit multiplier and the Multiplier and Accumulator in two different layouts. We assembly thousands of MTJ in rapid and clever way. The capability to simulate circuits composed of very large number of Nano devices avoiding the use of Magnetic Simulators: We simulate the whole multiplier in a few minutes. We are able to change the simulation model in order to have different levels of accuracy. Architectures for Nanocomputing 8 / 15

Assembly MTJ circuit in automatic way NANOcom is a Perl module that generates a VHDL crossbar circuit at Physical Level, taking in input: The physical level layout of the circuit (represented as matrix structure). The libraries describing: The crossbar VHDL component. The Nanotechnology devices to map in the crossbar. Return in output one VHDL file containing: The VHDL representation of the Crossbar structure (called Matrix). The VHDL configuration that bind the library component to the Matrix Cell. Architectures for Nanocomputing 9 / 15

Adopted Methodology Starting from a RT or Gate level circuit description, we can derive the MTJ implementation as follows: 1 Circuit decomposition in simple structures: AND, OR, NOT, MV, WIRE. 2 Crossbar layout type definition (Bound or Free MTJ clock zone definition) 3 Architectural Layout definition (placing input, output and inteconnnection). 4 Definition of consequent clock zone (Coloring, if not possible go to 2). 5 Simulation parameters measurements (Number of MTJ, circuit area, Latency...) Architectures for Nanocomputing 10 / 15

MTJ Structures definition 1 Define the ternary operator of Majority Voting (Low Effort in this technology) 2 Implement the basic AND/OR gates on the MV structure. 3 Define the behaviour of interconnection structures (MTJ Wires are intrinsecally pipelined). 3 Define a syncronization model using wl, bl and sl (MTJ Clock) Figure: AND, OR, NOT and WIREs Architectures for Nanocomputing 11 / 15

4 bit Multiplier - Layout Definition AREA: 54 rows 157 columns MTJs Number: 2217 MTJ % Occupied Area: 26.15% LATENCY: 17 MTJ clock cycle Architectures for Nanocomputing 12 / 15

Simulation SetUp 1 Generate simulation files with NANOcom. 2 Bind with the MTJ clocks generators. 3 Join with the MTJ components library. 4 Generate the stimuli and run the simulation. Architectures for Nanocomputing 13 / 15

Papers Published: "Composite Fields against Side Channel Analysisfor the Advanced Encryption Standard", Bollo, M., Maistri, P. (2014, December) accepted at 21st IEEE International Conference on Electronics Circuits & Systems (ICECS), 2014. "Physical Design and Testing of Nano Magnetic Architectures", Turvani, G., Tohti, A., Bollo, M., Riente, F., Vacca, M., Graziano, M., Zamboni, M. (2014, May). Design & Technology of Integrated Systems In Nanoscale Era (DTIS), 2014 9th IEEE International Conference On (pp. 1-6). IEEE. Architectures for Nanocomputing 14 / 15

Future Work Design and simulate a fully architectures for Advance Encryption Standard computation based on MTJ model. Study and design new nanoelectronic device models. Improve and generalize the NANOcom Module in order to obtain different technology layouts. Study and test the benefit of nanocomputing in Crypyographyc Accelerators. Architectures for Nanocomputing 15 / 15