LoopBuster Hardware Loop Detection in Fast Mesh Ethernet Networks Uriel Peled and Tal Kol Guided by Boaz Mizrahi Advised by Gideon Kaempfer Digital Systems.

Slides:



Advertisements
Similar presentations
NetFPGA Project: 4-Port Layer 2/3 Switch Ankur Singla Gene Juknevicius
Advertisements

Sumitha Ajith Saicharan Bandarupalli Mahesh Borgaonkar.
Data Protection Card Submit: Assaf Matia Technion Guide: Eran Segev Rafael Guide: Henri Delmar Winter & Spring 2004.
LoopBuster Hardware Loop Detection in Fast Mesh Ethernet Networks Uriel Peled and Tal Kol Guided by Boaz Mizrahi Advised by Gideon Kaempfer Digital Systems.
Characterization Presentation Neural Network Implementation On FPGA Supervisor: Chen Koren Maria Nemets Maxim Zavodchik
1 Performed By: Khaskin Luba Einhorn Raziel Einhorn Raziel Instructor: Rivkin Ina Spring 2004 Spring 2004 Virtex II-Pro Dynamical Test Application Part.
LoopBuster Hardware Loop Detection in Fast Mesh Ethernet Networks Uriel Peled and Tal Kol Guided by Boaz Mizrahi Advised by Gideon Kaempfer Digital Systems.
© 2004 Xilinx, Inc. All Rights Reserved Implemented by : Alon Ben Shalom Yoni Landau Project supervised by: Mony Orbach High speed digital systems laboratory.
LoopBuster Hardware Loop Detection in Fast Mesh Ethernet Networks Uriel Peled and Tal Kol Guided by Boaz Mizrahi Advised by Gideon Kaempfer Digital Systems.
Reliable Data Storage using Reed Solomon Code Supervised by: Isaschar (Zigi) Walter Performed by: Ilan Rosenfeld, Moshe Karl Spring 2004 Part A Final Presentation.
1 Performed by: Lin Ilia Khinich Fanny Instructor: Fiksman Eugene המעבדה למערכות ספרתיות מהירות High Speed Digital Systems Laboratory הטכניון - מכון טכנולוגי.
1 Multi - Core fast Communication for SoPC Multi - Core fast Communication for SoPC Technion – Israel Institute of Technology Department of Electrical.
VirtexIIPRO FPGA Device Functional Testing In Space environment. Performed by: Mati Musry, Yahav Bar Yosef Instuctor: Inna Rivkin Semester: Winter/Spring.
LoopBuster Hardware Loop Detection in Fast Mesh Ethernet Networks Uriel Peled and Tal Kol Guided by Boaz Mizrahi Advised by Gideon Kaempfer Digital Systems.
DSP Algorithm on System on Chip Performed by : Einat Tevel Supervisor : Isaschar Walter Accompanying engineers : Emilia Burlak, Golan Inbar Technion -
Configurable System-on-Chip: Xilinx EDK
הטכניון - מכון טכנולוגי לישראל הפקולטה להנדסת חשמל Technion - Israel institute of technology department of Electrical Engineering Virtex II-PRO Dynamical.
Reliable Data Storage using Reed Solomon Code Supervised by: Isaschar (Zigi) Walter Performed by: Ilan Rosenfeld, Moshe Karl Spring 2004 Midterm Presentation.
Technion Digital Lab Project Xilinx ML310 board based on VirtexII-PRO programmable device Students: Tsimerman Igor Firdman Leonid Firdman Leonid.
1 Fast Communication for Multi – Core SOPC Technion – Israel Institute of Technology Department of Electrical Engineering High Speed Digital Systems Lab.
1 FINAL PRESENTATION PART A Implementation of generic interface To electronic components via USB2 Connection Supervisor Daniel Alkalay System architectures.
Interface of DSP to Peripherals of PC Spring 2002 Supervisor: Broodney, Hen | Presenting: Yair Tshop Michael Behar בס " ד.
Ethernet Bomber Ethernet Packet Generator for network analysis Oren Novitzky & Rony Setter Advisor: Mony Orbach Started: Spring 2008 Part A final Presentation.
SSS 4/9/99CMU Reconfigurable Computing1 The CMU Reconfigurable Computing Project April 9, 1999 Mihai Budiu
Implementation of DSP Algorithm on SoC. Mid-Semester Presentation Student : Einat Tevel Supervisor : Isaschar Walter Accompaning engineer : Emilia Burlak.
הטכניון - מכון טכנולוגי לישראל הפקולטה להנדסת חשמל Technion - Israel institute of technology department of Electrical Engineering Virtex II-PRO Dynamical.
Technion Digital Lab Project Performance evaluation of Virtex-II-Pro embedded solution of Xilinx Students: Tsimerman Igor Firdman Leonid Firdman.
Implementation of DSP Algorithm on SoC. Characterization presentation Student : Einat Tevel Supervisor : Isaschar Walter Accompany engineer : Emilia Burlak.
v8.2 System Generator Audio Quick Start
System Architecture A Reconfigurable and Programmable Gigabit Network Interface Card Jeff Shafer, Hyong-Youb Kim, Paul Willmann, Dr. Scott Rixner Rice.
General Purpose FIFO on Virtex-6 FPGA ML605 board midterm presentation
Sept EE24C Digital Electronics Project Design of a Digital Alarm Clock.
Out-of-Order OpenRISC 2 semesters project Semester A: Implementation of OpenRISC on XUPV5 board Final A Presentation By: Vova Menis-Lurie Sonia Gershkovich.
Juanjo Noguera Xilinx Research Labs Dublin, Ireland Ahmed Al-Wattar Irwin O. Irwin O. Kennedy Alcatel-Lucent Dublin, Ireland.
Students: Oleg Korenev Eugene Reznik Supervisor: Rolf Hilgendorf
General Purpose FIFO on Virtex-6 FPGA ML605 board Students: Oleg Korenev Eugene Reznik Supervisor: Rolf Hilgendorf 1 Semester: spring 2012.
Digital signature using MD5 algorithm Hardware Acceleration
Out-of-Order OpenRISC 2 semesters project Semester A: Implementation of OpenRISC on XUPV5 board Midterm Presentation By: Vova Menis-Lurie Sonia Gershkovich.
Live Action First Person Shooter Game Patrick Judd Ian Katsuno Bao Le.
Elad Hadar Omer Norkin Supervisor: Mike Sumszyk Winter 2010/11, Single semester project. Date:22/4/12 Technion – Israel Institute of Technology Faculty.
1 WORLD CLASS – through people, technology and dedication High level modem development for Radio Link INF3430/4431 H2013.
High-Level Interconnect Architectures for FPGAs An investigation into network-based interconnect systems for existing and future FPGA architectures Nick.
GBT Interface Card for a Linux Computer Carson Teale 1.
Applied research laboratory David E. Taylor Users Guide: Fast IP Lookup (FIPL) in the FPX Gigabit Kits Workshop 1/2002.
High-Level Interconnect Architectures for FPGAs Nick Barrow-Williams.
VHDL Project Specification Naser Mohammadzadeh. Schedule  due date: Tir 18 th 2.
NIOS II Ethernet Communication Final Presentation
1 Abstract & Main Goal המעבדה למערכות ספרתיות מהירות High speed digital systems laboratory The focus of this project was the creation of an analyzing device.
By V. Koutsoumpos, C. Kachris, K. Manolopoulos, A. Belias NESTOR Institute – ICS FORTH Presented by: Kostas Manolopoulos.
Part A Presentation Implementation of DSP Algorithm on SoC Student : Einat Tevel Supervisor : Isaschar Walter Accompanying engineer : Emilia Burlak The.
Verification Methodology of Gigabit Switch System 1999/9/9 Yi Ju Hwan.
4/19/20021 TCPSplitter: A Reconfigurable Hardware Based TCP Flow Monitor David V. Schuehler.
Final Presentation Implementation of DSP Algorithm on SoC Student : Einat Tevel Supervisor : Isaschar Walter Accompanying engineer : Emilia Burlak The.
A Monte Carlo Simulation Accelerator using FPGA Devices Final Year project : LHW0304 Ng Kin Fung && Ng Kwok Tung Supervisor : Professor LEONG, Heng Wai.
Network On Chip Platform
IMPLEMENTATION OF MIPS 64 WITH VERILOG HARDWARE DESIGN LANGUAGE BY PRAMOD MENON CET520 S’03.
Technion - Israel institute of technology department of Electrical Engineering High speed digital systems laboratory 40Gbit Signal Generator for Ethernet.
Proposal for an Open Source Flash Failure Analysis Platform (FLAP) By Michael Tomer, Cory Shirts, SzeHsiang Harper, Jake Johns
DDRIII BASED GENERAL PURPOSE FIFO ON VIRTEX-6 FPGA ML605 BOARD PART B PRESENTATION STUDENTS: OLEG KORENEV EUGENE REZNIK SUPERVISOR: ROLF HILGENDORF 1 Semester:
Teaching Digital Logic courses with Altera Technology
Lab Environment and Miniproject Assignment Spring 2009 ECE554 Digital Engineering Laboratory.
Field Programmable Port Extender (FPX) 1 Modular Design Techniques for the Field Programmable Port Extender John Lockwood and David Taylor Washington University.
Internal Logic Analyzer Characterization presentation By: Moran Katz and Zvika Pery Mentor: Moshe Porian Dual-semester project Spring 2012.
ROD Activities at Dresden Andreas Glatte, Andreas Meyer, Andy Kielburg-Jeka, Arno Straessner LAr Electronics Upgrade Meeting – LAr Week September 2009.
Maj Jeffrey Falkinburg Room 2E46E
Lab 1: Using NIOS II processor for code execution on FPGA
Logic Analyzers EECS150 Fall Lab Lecture #5 Arjun Singh
The performance requirements for DSP applications continue to grow and the traditional solutions do not adequately address this new challenge Paradigm.
Logic Analyzers EECS150 Fall Lab Lecture #5 Arjun Singh
NetFPGA - an open network development platform
Presentation transcript:

LoopBuster Hardware Loop Detection in Fast Mesh Ethernet Networks Uriel Peled and Tal Kol Guided by Boaz Mizrahi Advised by Gideon Kaempfer Digital Systems Laboratory Faculty of Electrical Engineering, Technion Winter 2007 – Spring 2009 Completion Presentation

Ethernet Drawbacks Tree Topologies For Loop Prevention A B C

LoopBuster Stop Loops Without Tree Topology A B C New Hardware Device: “LoopBuster” Improved Switches: Changed Learning Improved Switches: Changed Learning Mesh Topology: Loops Allowed!

Design Challenges LoopBuster Device Support very high throughputs Ethernet supports 1Gbps and 10Gbps links Implementation must be in hardware Use limited amount of on-chip memory Naïve implementation requires 10Mbit for a single 10Gbps interface Minimal effect on hosting network Remove looping packets quickly Minimize false positives Rely on existing standards / network equipment

packet packet packet packet packet packet packet packet packet packet Filter Conceptual Diagram LoopBuster Device Low memory filters in decreasing size Still effective – packet rate decreases We pay with N+1 mandatory loops

Project Milestones   Full Network Software Simulation pre-hardware implementation   Analyze Algorithm Parameters   Design LoopBuster Device macro, micro architectures, Verilog implementation   Board Bring-up board selection, cores, GbE   Testing, Validation and Debug   Demonstrate Working Prototype

Algorithm Parameters Performance-Cost Tradeoffs Number of filter chains Parallel chains for different MLTTs Different chains for different traffic types Number of filters in each chain Number of loops before terminating a packet Memory size of each filter in each chain Minimize false positives while saving memory Filter scheduling and control patterns Lock / unlock states, reset filter memory

Network Simulation in Software Minimal HW (1PC), C++, SW Timeline Real-world rates, real-world traffic Based on a genetic algorithm Genetic representation: filter size list ( 13,12,10,10,9,9,8,8,7,7,6) Improve a pre-defined fitness function False positives over real traffic + Total memory size Two-stage mutation General (add/remove filter, change filter size, switch filters) Specific (num filters, size of largest, create descending chain) Algorithm Analysis Empirical Param Selection

Theoretical algorithm performance analysis Probability model for a filter / filter chain C collisions with X packets through an N-bit filter Occupancy problem, numerical solution in C++ Algorithm Analysis Theoretical Param Selection

Modular Filter Chain Design Traffic sensitive Code / Runtime Configurable Parameters UART controlled lock/unlock/reset patterns 2 Clock Domains 125 MHz (GbE), 31 MHz (Processing) Packet Pipeline Processing No store and forward (untraditional MAC) Hardware only (no Power PC) Preliminary Design Decision Algorithm Analysis Conclusions

Architecture Board Block Diagram

Architecture General Block Diagram

Packet Transceiver Block Diagram 125 MHz clock boundary Ethernet data in 8-bit units 31 MHz clock boundary Ethernet data in 32-bit units

LoopBuster Filter Block Diagram

LoopBuster Filter Implementation Supports two concurrent packet paths Unique clock domain: 125 MHz lb_filter_memory (Memory) Filter BRAM wrapper (2 asynchronous ports) Wide write port for asynchronous reset (FSM) Narrow read/write port for filter memory access Supports lock/unlock states lb_filter_state_machine (Filter Logic) Mutual exclusion for memory access Fine-grained locking (cycle requirement per state)

Board Selection Selected Board Memec FF1152 Xilinx Virtex-II Pro Existing in lab ($0) 2 SFP Modules 1Gbps Eth. RJ45 Gidi (~$200) PCS/PMA Core Required for SFP Free from Xilinx ($0)

Board Functionally Test Download a full working Ethernet example project to test UART, SFPs, LEDs, FPGA, cable correct operation Working DCM Synthesize a working DCM with 125Mhz, 31.5Mhz clock trees. Output main control signals to LEDs (ticker, locked) Core Linkup Test Configure Xilinx gig_eth_pcs_pma CORE for 1GE functionality with correct parameters, timing constrains and physical locs Output CORE status signals to detect linkup Loopback Tests Perform CORE loopback test, Packet transceiver loopback, full data path loopback GbE Board Bring-Up Vertical Development Stages

Behavioral Simulations Pre-synthesis Verilog for logic functionality on ModelSim Post-Route Timing Simulations Post-synthesis Verilog for timing on Xilinx ISE 9.2 Automatic Simulation Test-bench Script-based scenario test-benches for core modules (like pt_rx) Automated ModelSim with debug textual log file ($fdisplay) On-Board Live Debugging Status signals to LED, R/W of debug registers with UART Custom event-based debug code (output to UART) System Integration Testing Stream raw Ethernet traffic through device, Packetyzer sniffer Testing and Validation

Planned / Actual Schedule Full Network Software Simulation PLAN: 2 months ACTUAL: 2 months Analyze Algorithm Parameters PLAN: 1 month ACTUAL: 1 month Design LoopBuster Device (micro-macro) PLAN: 3 months ACTUAL: 3 months Board Bring-up PLAN: 1 month ACTUAL: 3 months Testing, Validation and Debug PLAN: 2 months ACTUAL: 4 months Demonstrate Working Prototype PLAN: 1 month ACTUAL: 1 month TOTAL PLAN: 10 months ACTUAL: 14 months

Main Project Achievements Algorithm analysis and params based on software simulations Working LoopBuster prototype in hardware (FPGA) Testing and validation environment LoopBuster-PC communication and control platform Further Work LBP switch implementation with revised learning algorithm Complete network solution demonstration in hardware In-depth LoopBuster algorithm analysis and optimization with hardware-based results Achievements and Further Work