Designing a Fast and Reliable Memory with Memristor Technology

Slides:



Advertisements
Similar presentations
30 nm © 2005 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Atomic Switch ITRS Emerging.
Advertisements

Handling Resistance Drift in Phase Change Memory - Device, Circuit, Architecture, and System Solutions Manu Awasthi, Manjunath Shevgoor, Kshitij Sudan,
RAM (RANDOM ACCESS MEMORY)
Introduction to the TRAMS project objectives and results in Y1 Antonio Rubio, Ramon Canal UPC, Project coordinator CASTNESS’11 WORKSHOP ON TERACOMP FET.
COEN 180 SRAM. High-speed Low capacity Expensive Large chip area. Continuous power use to maintain storage Technology used for making MM caches.
Computer Organization and Architecture
Computer Organization and Architecture
University of Michigan Electrical Engineering and Computer Science University of Michigan Electrical Engineering and Computer Science University of Michigan.
+ CS 325: CS Hardware and Software Organization and Architecture Internal Memory.
5-1 Memory System. Logical Memory Map. Each location size is one byte (Byte Addressable) Logical Memory Map. Each location size is one byte (Byte Addressable)
Rethinking DRAM Design and Organization for Energy-Constrained Multi-Cores Aniruddha N. Udipi, Naveen Muralimanohar*, Niladrish Chatterjee, Rajeev Balasubramonian,
Sistemi Elettronici Programmabili1 Progettazione di circuiti e sistemi VLSI Anno Accademico Lezione Memorie (vedi anche i file pcs1_memorie.pdf.
Elettronica T AA Digital Integrated Circuits © Prentice Hall 2003 SRAM & DRAM.
Probabilistic Design Methodology to Improve Run- time Stability and Performance of STT-RAM Caches Xiuyuan Bi (1), Zhenyu Sun (1), Hai Li (1) and Wenqing.
EE314 Basic EE II Review: Kirchhoff’s Laws and Resistive Circuits.
Reducing Read Latency of Phase Change Memory via Early Read and Turbo Read Feb 9 th 2015 HPCA-21 San Francisco, USA Prashant Nair - Georgia Tech Chiachen.
Nak Hee Seong Sungkap Yeo Hsien-Hsin S. Lee
Twin Logic Gates – Improved Logic Reliability by Redundancy concerning Gate Oxide Breakdown Hagen Sämrow, Claas Cornelius, Frank Sill, Andreas Tockhorn,
Overcoming the Challenges of Crossbar Resistive Memory Architecture
Kevin Walsh CS 3410, Spring 2010 Computer Science Cornell University Memory See: P&H Appendix C.8, C.9.
Hakim Weatherspoon CS 3410, Spring 2011 Computer Science Cornell University Memory See: P&H Appendix C.8, C.9.
11/29/2004EE 42 fall 2004 lecture 371 Lecture #37: Memory Last lecture: –Transmission line equations –Reflections and termination –High frequency measurements.
June 20 th 2004University of Utah1 Microarchitectural Techniques to Reduce Interconnect Power in Clustered Processors Karthik Ramani Naveen Muralimanohar.
Die-Hard SRAM Design Using Per-Column Timing Tracking
Chapter 5 Internal Memory
Feb 14 th 2005University of Utah1 Microarchitectural Wire Management for Performance and Power in partitioned architectures Rajeev Balasubramonian Naveen.
University of Utah 1 The Effect of Interconnect Design on the Performance of Large L2 Caches Naveen Muralimanohar Rajeev Balasubramonian.
Feb 14 th 2005University of Utah1 Microarchitectural Wire Management for Performance and Power in Partitioned Architectures Rajeev Balasubramonian Naveen.
Low Voltage Low Power Dram
QUANTIFYING THE RELATIONSHIP BETWEEN THE POWER DELIVERY NETWORK AND ARCHITECTURAL POLICIES IN A 3D-STACKED MEMORY DEVICE Manjunath Shevgoor, Niladrish.
1 Towards Scalable and Energy-Efficient Memory System Architectures Rajeev Balasubramonian School of Computing University of Utah.
©Wen-mei W. Hwu and David Kirk/NVIDIA, ECE408/CS483/ECE498AL, University of Illinois, ECE408/CS483 Applied Parallel Programming Lecture 7: DRAM.
Leveraging Heterogeneity in DRAM Main Memories to Accelerate Critical Word Access Niladrish Chatterjee Manjunath Shevgoor Rajeev Balasubramonian Al Davis.
UNDERSTANDING THE ROLE OF THE POWER DELIVERY NETWORK IN 3-D STACKED MEMORY DEVICES Manjunath Shevgoor, Niladrish Chatterjee, Rajeev Balasubramonian, Al.
Memristor – The Fourth Fundamental Circuit Element
1 University of Utah & HP Labs 1 Optimizing NUCA Organizations and Wiring Alternatives for Large Caches with CACTI 6.0 Naveen Muralimanohar Rajeev Balasubramonian.
TOWARDS AN EARLY DESIGN SPACE EXPLORATION TOOL SET FOR STT-RAM DESIGN Philip Asare and Ben Melton.
Dynamic Power Noise Analysis Method for Memory Designs
Coding and Algorithms for Memories Lecture 9 1.
Contents:  Introduction  what do you mean by memristor.  Need for memristor.  The types of memristor.  Characteristics of memristor.  The working.
CPEN Digital System Design
Project 1: DRAM timing violation due to PV Due to PV, transistor and capacitor may have variations in their dimensions, causing charging time of a cell.
Low-Power SRAM ECE 4332 Fall 2010 Team 2: Yanran Chen Cary Converse Chenqian Gan David Moore.
Memristor  Naveed Bashir  Enroll no. 38/09  Roll no. 06  7 th Semester  Department of Electrical Engg., NIT Srinagar.
By Edward A. Lee, J.Reineke, I.Liu, H.D.Patel, S.Kim
ECE Memristors for computing & Future Non-Volatile Memory Application- Beyond transistors and silicon technology Somnath Chakraborty, Dept. Of ECE, UMASS.
Feb 14 th 2005University of Utah1 Microarchitectural Wire Management for Performance and Power in Partitioned Architectures Rajeev Balasubramonian Naveen.
Computer Architecture Lecture 24 Fasih ur Rehman.
Copyright © 2010 Houman Homayoun Houman Homayoun National Science Foundation Computing Innovation Fellow Department of Computer Science University of California.
Enabling Big Memory with Emerging Technologies Manjunath Shevgoor Enabling Big Memory with Emerging Technologies1.
Carnegie Mellon University, *Seagate Technology
A High-Speed & High-Capacity Single-Chip Copper Crossbar John Damiano, Bruce Duewer, Alan Glaser, Toby Schaffer, John Wilson, and Paul Franzon North Carolina.
1 November 11, 2015 The Evolution of Memory Architecture November 11, 2015 Dr. Amit Berman Senior Engineer, Samsung.
1 Lecture 20: Big Data, Memristors Today: architectures for big data, memristors.
Efficient Scrub Mechanisms for Error-Prone Emerging Memories Manu Awasthi ǂ, Manjunath Shevgoor⁺, Kshitij Sudan⁺, Rajeev Balasubramonian⁺, Bipin Rajendran.
33 rd IEEE International Conference on Computer Design ICCD rd IEEE International Conference on Computer Design ICCD 2015 Improving Memristor Memory.
Coding and Algorithms for Memories Lecture 7 1.
MEMRISTOR The Fourth Fundamental Circuit Element.
Modeling of Failure Probability and Statistical Design of Spin-Torque Transfer MRAM (STT MRAM) Array for Yield Enhancement Jing Li, Charles Augustine,
William Stallings Computer Organization and Architecture 7th Edition
MEMRISTOR PRESENTED BY Varun Thomas S7 R No 39.
PRESENTED BY SAI KRISHNA.R (2-1) NRIIT TEJASWI.K(2-1)
William Stallings Computer Organization and Architecture 7th Edition
BIC 10503: COMPUTER ARCHITECTURE
Memory.
Aniruddha N. Udipi, Naveen Muralimanohar*, Niladrish Chatterjee,
Manjunath Shevgoor, Rajeev Balasubramonian, University of Utah
STT-RAM Design Fengbo Ren Advisor: Prof. Dejan Marković Dec. 3rd, 2010
Presentation transcript:

Designing a Fast and Reliable Memory with Memristor Technology Manjunath Shevgoor, Rajeev Balasubramonian, Naveen Muralimanohar University of Utah, HP Labs Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Background Store data in the form of resistance Metal oxide sandwiched between two electrodes Inherently non conducting Creation of conductive Filaments of oxygen vacancies reduces resistance Source: Cong Xu et al., Modeling and Design Analysis of 3D Vertical Resistive Memory - A Low Cost Cross-Point Architecture, ASPDAC 2014 Designing a Fast and Reliable Memory with Memristor Technology

Voltage Dependent Resistance The resistance of a ReRAM cell is not constant but varies with the applied voltage Kr(p, V ) = p * R(V/p)/R(V ) R(V/p) and R(V ) are the equivalent resistance of the cell biased at V/p and V Kr is the non-linearity. Eg: if Kr=20, resistance increases 10x when voltage is halved Resistance decreases with increasing voltage Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Bit Line Word DRAM Cell Word Line Bit Memristor Cell Bit Line Word PCM Cell Cell Size of 4F2 Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Cross Point Structure Driver Transistors Selected Cell Mem- ristor Selector Memristor Cell Because of non-linearity, it is possible to select a cell without an access transistor. Arrays can be layered vertically without resorting to 3D stacking. Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Reading and Writing 0V Vdd/2 Selected Cell Sneak Current Vdd Half Selected Cells Driver Transistors Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Bit Lines RW RW RW V Word Lines VW1 VW2 VWN RW RW RW VWN1 RW RW RW V/2 RW RW RW VWNM V/2 V/2 Bit Line Mux Bit line and word line resistances eat into the cell Voltage Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Effects of Ileak Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Effects of Ileak Decreases Voltage at selected cell Increases Write Latency Can cause Write Failure Distorts bit line current Increases read complexity Decreases read margin Designing a Fast and Reliable Memory with Memristor Technology

Sneak path currents can distort Iread Vread Iread Vread/2 Ileak Vread/2 Ileak Vread/2 Ileak Vread/2 Vread/2 Vread/2 Sneak path currents can distort Iread Designing a Fast and Reliable Memory with Memristor Technology

Step 1: Read background current (Ileak) Vread/2 Ileak Vread/2 Ileak Vread/2 Ileak Vread/2 Ileak Vread/2 Vread/2 Vread/2 Step 1: Read background current (Ileak) Designing a Fast and Reliable Memory with Memristor Technology

Step 2: Read total Vread current (Iread) Ileak Vread/2 Ileak Vread/2 Ileak Vread/2 Vread/2 Vread/2 Step 2: Read total Vread current (Iread) Designing a Fast and Reliable Memory with Memristor Technology

State of selected cell determines Iread ~ Ileak tBG_READ tREAD Read Latency Designing a Fast and Reliable Memory with Memristor Technology

Proposal 1: Re-use value in sample and hold circuit Vread Vread/2 Pprech Pacc Vr S2 Sensing Circuit S1 Sample and Hold Sneak Current Proposal 1: Re-use value in sample and hold circuit Designing a Fast and Reliable Memory with Memristor Technology

Reusing Sneak Current Read Rows Sneak Current uA Columns Designing a Fast and Reliable Memory with Memristor Technology

Re-Use Sneak Current Reading for the same Column tBG_READ tREAD tREAD Read Latency1 Read Latency2 Designing a Fast and Reliable Memory with Memristor Technology

Impact of Cell Location Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Word Line Drivers Bit Line Mux Longer write latencies Increased error rates Designing a Fast and Reliable Memory with Memristor Technology

Array 1 Array 2 Array 3 Array 512 Bit 1 Bit 2 Bit 3 Bit 512 64 Byte Cache line Default Mapping Leads to some lines with high error rate Designing a Fast and Reliable Memory with Memristor Technology

Proposal 2: Stagger the array mapping Cacheline 1 Cacheline 2 Cacheline 3 Cacheline 4 1 2 3 Nth bit in cacheline Array 0 Array 1 Array 2 Array 3 Default Mapping 1 3 2 Proposed Mapping Designing a Fast and Reliable Memory with Memristor Technology

Proposal 2: Impact on Write Latency 1 2 3 Nth bit in cacheline Array 0 Array 1 Array 2 Array 3 Default Mapping 1 3 2 Proposed Mapping Designing a Fast and Reliable Memory with Memristor Technology

Proposal 3: Compress to reduce write latency Array 1 Array 2 Array 3 Array 512 Bit 1 Bit 2 Bit 3 Bit 512 64 Byte Cache line 1 3 2 Proposed Mapping With 50% Compression Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Conclusions With great density come a few challenges Sneak Currents limit array size, complicate reads, and delay writes Affect reliability Background current can be reused Reliability can be improved at the cost of write latency Compression can reduce write latency 13% performance improvement 30X reduction in multi bit error probability Work in progress Designing a Fast and Reliable Memory with Memristor Technology

Designing a Fast and Reliable Memory with Memristor Technology Thank You Designing a Fast and Reliable Memory with Memristor Technology