Supporting Runtime Reconfiguration on Network Processors Kevin Lee Lancaster University

Slides:



Advertisements
Similar presentations
Network II.5 simulator ..
Advertisements

Operating Systems Components of OS
PhD 2 nd year panel Kevin lee October 2004 A Generic Programming Model for Network Processors Part Deux.
Supercharging PlanetLab : a high performance, Multi-Application, Overlay Network Platform Written by Jon Turner and 11 fellows. Presented by Benjamin Chervet.
Department of Computer Science and Engineering University of Washington Brian N. Bershad, Stefan Savage, Przemyslaw Pardyak, Emin Gun Sirer, Marc E. Fiuczynski,
A Software Defined Approach to Unified IPv6 Transition.
Chapter 8 Hardware Conventional Computer Hardware Architecture.
Approaches to EJB Replication. Overview J2EE architecture –EJB, components, services Replication –Clustering, container, application Conclusions –Advantages.
Extensible Networking Platform IWAN 2005 Extensible Network Configuration and Communication Framework Todd Sproull and John Lockwood
Page 1 Building Reliable Component-based Systems Chapter 16 - Component based embedded systems Chapter 16 Component based embedded systems.
Contiki A Lightweight and Flexible Operating System for Tiny Networked Sensors Presented by: Jeremy Schiff.
1 Router Construction II Outline Network Processors Adding Extensions Scheduling Cycles.
NodeOS Programming Forwarding Functions for Extensible Routers Department of Computer Science Princeton University Router Workshop Yitzchak.
Performance Analysis of the IXP1200 Network Processor Rajesh Krishna Balan and Urs Hengartner.
The Future of Middleware R&D Geoff Coulson Distributed Multimedia Research Group Computing Dept, Lancaster University
Managing Agent Platforms with the Simple Network Management Protocol Brian Remick Thesis Defense June 26, 2015.
V1.00 © 2009 Research In Motion Limited Introduction to Mobile Device Web Development Trainer name Date.
Embedded Transport Acceleration Intel Xeon Processor as a Packet Processing Engine Abhishek Mitra Professor: Dr. Bhuyan.
Router Construction II Outline Network Processors Adding Extensions Scheduling Cycles.
Copyright Arshi Khan1 System Programming Instructor Arshi Khan.
1 A survey on Reconfigurable Computing for Signal Processing Applications Anne Pratoomtong Spring2002.
Mahesh Wagh Intel Corporation Member, PCIe Protocol Workgroup.
Checkpoint & Restart for Distributed Components in XCAT3 Sriram Krishnan* Indiana University, San Diego Supercomputer Center & Dennis Gannon Indiana University.
Lecture 18 Lecture 18: Case Study of SoC Design ECE 412: Microcomputer Laboratory.
LiNK: An Operating System Architecture for Network Processors Steve Muir, Jonathan Smith Princeton University, University of Pennsylvania
Extensibility, Safety and Performance in the SPIN Operating System Ashwini Kulkarni Operating Systems Winter 2006.
Improving Network I/O Virtualization for Cloud Computing.
Dynamic Reconfiguration Dynamic selection of handler functionality: currently through use of parameterizable handlers or by selecting from a set of existing.
1 Liquid Software Larry Peterson Princeton University John Hartman University of Arizona
Architecture Examples And Hierarchy Samuel Njoroge.
To be smart or not to be? Siva Subramanian Polaris R&D Lab, RTP Tal Lavian OPENET Lab, Santa Clara.
Heavy and lightweight dynamic network services: challenges and experiments for designing intelligent solutions in evolvable next generation networks Laurent.
Open Service Gateway Initiative (OSGi) Reporter : 林學灝 侯承育 1.
Constructing Services with Interposable Virtual Hardware Author: Andrew Whitaker, Richard S. Cox, Marianne Shaw, and Steven D. Gribble Presenter: Huajing.
ISO/IEC SC 25/WG 1 ISO/IEC : CodeBase Tag Discussion Ron Ambrosio IBM TJ Watson Research Center.
Performance Characterization and Architecture Exploration of PicoRadio Data Link Layer Mei Xu and Rahul Shah EE249 Project Fall 2001 Mentor: Roberto Passerone.
Tool Integration with Data and Computation Grid GWE - “Grid Wizard Enterprise”
TAL7011 – Lecture 4 UML for Architecture Modeling.
The Cosmic Cube Charles L. Seitz Presented By: Jason D. Robey 2 APR 03.
Welcome Experiences in the Use of MDA and UML in Developing NATO Standards 16 July 2008 Chris Raistrick, Kennedy KC.COM.
Distribution and components. 2 What is the problem? Enterprise computing is Large scale & complex: It supports large scale and complex organisations Spanning.
An Architecture and Prototype Implementation for TCP/IP Hardware Support Mirko Benz Dresden University of Technology, Germany TERENA 2001.
INNOV-10 Progress® Event Engine™ Technical Overview Prashant Thumma Principal Software Engineer.
1 Choices “Our object-oriented system architecture embodies the notion of customizing operating systems to tailor them to support particular hardware configuration.
EXTENSIBILITY, SAFETY AND PERFORMANCE IN THE SPIN OPERATING SYSTEM
Viktor Prasanna,Yogesh Simmhan, Alok Kumbhare, Sreedhar Natarajan 04/20/2012.
A Systematic Approach to the Design of Distributed Wearable Systems Urs Anliker, Jan Beutel, Matthias Dyer, Rolf Enzler, Paul Lukowicz Computer Engineering.
ND The research group on Networks & Distributed systems.
Intel Research & Development ETA: Experience with an IA processor as a Packet Processing Engine HP Labs Computer Systems Colloquium August 2003 Greg Regnier.
VMware vSphere Configuration and Management v6
Brian N. Bershad, Thomas E. Anderson, Edward D. Lazowska, and Henry M. Levy. Presented by: Tim Fleck.
Architecture Analysis and Design Language: An Overview Drew Gardner.
Network Virtualization Sandip Chakraborty. In routing table we keep both the next hop IP (gateway) as well as the default interface. Why do we require.
ECE 526 – Network Processing Systems Design Programming Model Chapter 21: D. E. Comer.
Software Systems Division (TEC-SW) ASSERT process & toolchain Maxime Perrotin, ESA.
Addressing Data Compatibility on Programmable Network Platforms Ada Gavrilovska, Karsten Schwan College of Computing Georgia Tech.
PhD VIVA Kevin lee 28 th July 2006 “OpenNP: A Generic Programming Model for Network Processors” Brief Introduction.
2014 Redefining the Data Center: White-Box Networking Jennifer Casella October 9, 2014 #GHC
Atrium Router Project Proposal Subhas Mondal, Manoj Nair, Subhash Singh.
Sensor network routing protocol for underground robot remote control Demonstration picture (IDF)
Heterogeneous Processing KYLE ADAMSKI. Overview What is heterogeneous processing? Why it is necessary Issues with heterogeneity CPU’s vs. GPU’s Heterogeneous.
1 Dynamic Classification in a Silicon-Based Forwarding Engine Technology Center, Nortel Networks & The University of Maryland Rob Jaeger
Introduction to Operating Systems Concepts
GPUNFV: a GPU-Accelerated NFV System
University of Maryland College Park
USB The topics covered, in order, are USB background
Dynamo: A Runtime Codesign Environment
Chapter 3: Windows7 Part 1.
Data Path through host/ANP.
Integrating Active Networking and Commercial-Grade Routing Platforms
Presentation transcript:

Supporting Runtime Reconfiguration on Network Processors Kevin Lee Lancaster University

Kevin Lee AINA th April Overview 1. Network Processors 2. Motivation 3. Our Approach 4. Implementation 5. Runtime Reconfiguration Support 6. Future Work 7. Summary/Questions

Kevin Lee AINA th April 1. Network Processors Designed to fill a gap in network technology Highly programmable Gigabit speed-routing Specialised packet processors

Kevin Lee AINA th April 1. The Intel IXP2400 Our Target Platform Main points: MicroEngines General Purpose CPU Multiple memory types Internal gigabit bus Programmable Hardware components Gigabit Ethernet ports

Kevin Lee AINA th April 2. Motivation Network Processors are inherently capable of Runtime Reconfiguration Lack of Runtime Reconfiguration support in Network Processor Software Support Active Networks Software Updates Supporting unforeseen events Load-balancing Reconfiguration of forwarding functions

Kevin Lee AINA th April 3. The Approach Using a Component Model: OpenCOM fine-grained and language-independent small run-time (~27K); simple and efficient components are primitive units of encapsulation and deployment deployed into capsules (for “capsule”, think container rather than active packet!) emphasises third-party composition and dynamic reconfiguration Support for Reflection using meta interfaces Architecture, Interception, Resources, Interface

Kevin Lee AINA th April 3. The Approach cont… OpenCOM enhancements for NPs Pluggable loaders To allow code to be loaded on third party processors Pluggable binders… To allow code to be bound within and between different processors Support for communication channels To allow proxy loading and binding

Kevin Lee AINA th April 3. OpenCOM Kernel API Implements the basic functionalities to support the creation and connection of OpenCOM components Loader API Allows use of loader plug-ins to load component types within specific caplets Binder API Allows use of binder plug-ins to bind specific component types together

Kevin Lee AINA th April 3. OpenCOM cont.

Kevin Lee AINA th April 3. In Practice

Kevin Lee AINA th April 4. OpenCOM Implementation on the Intel IXP2400 Central Runtime on XScale Load/Bind component natively Load plug-ins for third party Loading/Binding Microengine Loader Intra-Microengine Binder Inter-Microengine Binder PC-ARM Comms channel To allow loading/binding from PC/network to IXP

Kevin Lee AINA th April 4. OpenCOM Implementation on the Intel IXP2400 Source Code Diagram

Kevin Lee AINA th April 5. Runtime Reconfiguration on the IXP2400 To perform a reconfiguration Destroy a binding Load and instantiate a new component Bind the components together Implementation complexity hidden from programmer OpenCOM handles pausing of the packet flow Halting of the Microengines Control store is patched Microengines restarted Packet flow resumed

Kevin Lee AINA th April 5. Runtime Reconfiguration on the IXP2400 1: Dynamically Introduced Transcoding Service for the IXP2400

Kevin Lee AINA th April 5. Runtime Reconfiguration on the IXP2400 2: Load Balancing of the Microengines on the IXP2400

Kevin Lee AINA th April 5. Performance Reconfiguration takes 60ms per Microengine Throughput and Latency Measurements of OpenCOM components on the IXP2400 Microengines

Kevin Lee AINA th April 6. Future Work Further IXP2400 Implementation Componentise co-processors and network ports Loaders  Support these devices using OpenCOM runtime Binders  Between these components and microengine components Implementation on other platforms Current implementations IXP1200/IXP2400 Further platforms with different architecture To prove generality of the approach

Kevin Lee AINA th April Summary Network Processors are capable of supporting runtime reconfiguration Need for software support to enable this OpenCOM and extensions provides a potential solution Some overhead in providing this support Contact for further info: Questions???