MacroNET Network Topology

Slides:

Advertisements

Similar presentations

J. Kaiser University of Ulm Dept. Of Comp. Structures Jörg Kaiser Dept. Of Computer Structures University of Ulm And gets.

Advertisements

Comparison Of Network On Chip Topologies Ahmet Salih BÜYÜKKAYHAN Fall.

Interconnection Networks: Flow Control and Microarchitecture.

Dynamic Topology Optimization for Supercomputer Interconnection Networks Layer-1 (L1) switch –Dumb switch, Electronic “patch panel” –Establishes hard links.

Cache Coherent Distributed Shared Memory. Motivations Small processor count –SMP machines –Single shared memory with multiple processors interconnected.

High Performance Router Architectures for Network- based Computing By Dr. Timothy Mark Pinkston University of South California Computer Engineering Division.

1 Lecture 24: Interconnection Networks Topics: communication latency, centralized and decentralized switches (Sections 8.1 – 8.5)

Performance and Power Efficient On-Chip Communication Using Adaptive Virtual Point-to-Point Connections M. Modarressi, H. Sarbazi-Azad, and A. Tavakkol.

Asis AdvancedTCA Class What is a Backplane? A backplane is an electronic circuit board Sometimes called PCB (Printed Circuit Board) containing circuitry.

High-Performance Networks for Dataflow Architectures Pravin Bhat Andrew Putnam.

Extreme-scale computing systems – High performance computing systems Current No. 1 supercomputer Tianhe-2 at petaflops Pushing toward exa-scale computing.

Improving Capacity and Flexibility of Wireless Mesh Networks by Interface Switching Yunxia Feng, Minglu Li and Min-You Wu Presented by: Yunxia Feng Dept.

1 CHAPTER 8 TELECOMMUNICATIONSANDNETWORKS. 2 TELECOMMUNICATIONS Telecommunications: Communication of all types of information, including digital data,

Department of Computer Science and Engineering The Pennsylvania State University Akbar Sharifi, Emre Kultursay, Mahmut Kandemir and Chita R. Das Addressing.

Physical Topology Physical layout of the network nodes – Broad description of the network: no detail about device types, connection methods, addressing,...

Representing Uncertainty, Unknowns and Dynamics of Social Networks NX-Workshop on Social Network Analysis and Visualization for Public Safety 18 – 19 October.

Interconnect Networks Basics. Generic parallel/distributed system architecture On-chip interconnects (manycore processor) Off-chip interconnects (clusters.

Parallel Processing Presented by: Wanki Ho CS147, Section 1.

1 VLM 2 : A Very Lightweight Mobile Multicast System For Wireless Sensor Networks Anmol Sheth, Brian Shucker and Richard Han University of Colorado, Department.

Chapter 1 : Computer Networks.

Effective bandwidth with link pipelining Pipeline the flight and transmission of packets over the links Overlap the sending overhead with the transport.

PATH DIVERSITY WITH FORWARD ERROR CORRECTION SYSTEM FOR PACKET SWITCHED NETWORKS Thinh Nguyen and Avideh Zakhor IEEE INFOCOM 2003.

Lecture 13 Parallel Processing. 2 What is Parallel Computing? Traditionally software has been written for serial computation. Parallel computing is the.

1 Packet Switching Outline Switching and Forwarding Bridges and Extended LANs.

A seminar Presentation on NETWORK- ON- CHIP ARCHITECTURE EXPLORATION FRAMEWORK Under the supervision of Presented by Mr.G.Naresh,M.Tech., V.Sairamya Asst.

Data and Computer Communications Chapter 7 Circuit Switching and Packet Switching.

Chapter 1 Introduction.

Chapter 3 Part 3 Switching and Bridging

Overview Parallel Processing Pipelining

Parallel Architecture

Computer Network Collection of computers and devices connected by communications channels that facilitates communications among users and allows users.

Computer Network Topology

Semester 4 - Chapter 3 – WAN Design

Dynamic connection system

Lecture 23: Interconnection Networks

Connection System Serve on mutual connection processors and memory .

The Echo Algorithm The echo algorithm can be used to collect and disperse information in a distributed system It was originally designed for learning network.

Input Device :keyboard Types of key switch Components of Keyboard Keyboard switches Keyboard processor or circuitry Key matrix Types of key Switches.

Technology advancement in computer architecture

Lecture 18: Coherence and Synchronization

Architecture & Organization 1

Azeddien M. Sllame, Amani Hasan Abdelkader

Shuffle Exchange Point-to-point, Bus, Ring, Star, Mesh

Chapter 3 Part 3 Switching and Bridging

Network-on-Chip & NoCSim

Coordination and computation over wireless networks

MacroNET Lightnode Architecture & Network

Architecture & Organization 1

Static Interconnection Networks

On-time Network On-chip

Communication Networks NETW 501

Nodes, Branches, and Loops

Networked Real-Time Systems: Routing and Scheduling

Unique Features of Sensor Networks

Data Communications and Networking

RECONFIGURABLE NETWORK ON CHIP ARCHITECTURE FOR AEROSPACE APPLICATIONS

Chapter 3 Part 3 Switching and Bridging

Electronics for Physicists

Dhruv Gupta EEC 273 class project Prof. Chen-Nee Chuah

Wide Area Networks (WANs), Routing, and Shortest Paths

Optical communications & networking - an Overview

Static Interconnection Networks

Wide Area Networks (WANs), Routing, and Shortest Paths

Lecture 19: Coherence and Synchronization

Topology What are the consequences of using fewer than n2 connections?

Lecture 18: Coherence and Synchronization

Horizon: Balancing TCP over multiple paths in wireless mesh networks

Multiprocessors and Multi-computers

Unique Features of Sensor Networks

Chapter 2 from ``Introduction to Parallel Computing'',

Presentation transcript:

MacroNET Network Topology ELE-580i PRESENTATION-I 03/13/2003 Canturk ISCI

Introduction MacroNET: Drawbacks: Large area electronics + flexible & deformable <rolled like a window shade> Interactive Map << E-ink (www.eink.com) PowerPaper (www.powerpaper.com) Distributed processing units (light) Interconnect fabric (light) Drawbacks: A-Si over plastic (due to size and flexibility) Low n No p-device No ~CMOS Area ~ x400 of CMOS (~x2500 in Princeton) Currently O(10) T circuits 1/2/2019

PROJECT Definition: Motivation: Related Work: Sparse, Lightweight, Low node degree topology Transistor Budget Direct Network Compute & Route same + buffer=regfile/memory Some extent of path diversity Fault tolerance Motivation: Particular design challenges Tight complexity, fault tolerance Relaxed Latency, performance Related Work: More performance driven Lightweight networks – for on chip processors focus on reducing latency 1/2/2019

Design Cycle Transistor Budget Topology Routing & Compute Node Break the loop: Assume an initial Topology Define compute node Define the nature of Communication and instructions among nodes Acquire a representative Traffic pattern Redefine an appropriate topology Sparse Torus?? Move to CPU modeling: Detailed description (~dataflow) of CPU Back of the envelope Transistor Count Computation Anticipated Constraints: 1000 Ts per CPU Node degree <3 (connectivity??) Transistor Budget Topology Routing & Flow Control Traffic Pattern Compute Node Model Instructions & Communication 1/2/2019

Initialization Topology: mesh (degree = 4!) CPU instructions: Zoom around node(r,c) Dim / brighten Maketop N/S/W/E ( like rotate) Compute distance ** (Directed Routing) Show Places of interest, etc.?? Input style: from sensors Communication – Routing: Instruction dependent Channel Capacity: Infinite (for traffic pattern generation) This setup is very relaxed for traffic pattern generation!! 1/2/2019

CURRENT Status (r,c) (0,0) (1,3) (0,2) (3,3) (1,0) (0,1) (1,2) (3,0) Implementing MacroSIM Need to discuss communications… I.e. zoom (2,1): (r,c) (0,0) (1,3) (0,2) (3,3) (1,0) (0,1) (1,2) (3,0) (2,3) (3,2) (0,0) (0,1) (0,2) (0,3) (0,3) Initial Processor (r,c) (1,0) (1,1) (2,0) (3,1) (2,2) (1,1) (1,2) (1,3) Updated Processor Sent Instruction (2,0) (2,1) (2,1) (2,2) (2,3) Sent ACK Sent ACKed (3,0) (3,1) (3,2) (3,3) Used Channel 1/2/2019