Time Triggered Networks: use in space 2015 CCSDS spring SOIS Plenary 23 March 2015 Glenn Rakow/NASA-GSFC.

Slides:



Advertisements
Similar presentations
Distributed Systems Major Design Issues Presented by: Christopher Hector CS8320 – Advanced Operating Systems Spring 2007 – Section 2.6 Presentation Dr.
Advertisements

Bus Architectures for Satety- Critical Embedded Systems --by Harit Desai.
Multiple Processor Systems
1 The Time-Triggered Model of Computation Lior Zimet.
Chapter 14: Design Method --- data and architectural design Design -- A multistep process in which representations of data structure, program structure,
City University London
CS599 Software Engineering for Embedded Systems1 Software Engineering for Real-Time: A Roadmap Presentation by: Mandar Samant Raghbir Singh Banwait.
Design of Fault Tolerant Data Flow in Ptolemy II Mark McKelvin EE290 N, Fall 2004 Final Project.
Using Architecture Frameworks
EEC-681/781 Distributed Computing Systems Lecture 3 Wenbing Zhao Department of Electrical and Computer Engineering Cleveland State University
TTP ® - As Predictable as Time 1 Copyright © 2002, TTTech Computertechnik AG. All rights reserved. TTTech Powernode As a Linux Prototyping Platform for.
CprE 458/558: Real-Time Systems
ESA UNCLASSIFIED – For Official Use Deterministic Communication with SpaceWire Martin Suess CCSDS Spring Meeting /03/2015.
The Rare Glitch Project: Verifying Bus Protocols for Embedded Systems Edmund Clarke, Daniel Kroening Carnegie Mellon University.
SWE Introduction to Software Engineering
IV&V Facility Model-based Design Verification IVV Annual Workshop September, 2009 Tom Hempler.
©Ian Sommerville 2004Software Engineering, 7th edition. Chapter 12 Slide 1 Distributed Systems Design 1.
Network Topologies.
Chapter 10 Architectural Design
PMIT-6102 Advanced Database Systems
IMS 4212: Distributed Databases 1 Dr. Lawrence West, Management Dept., University of Central Florida Distributed Databases Business needs.
S1.6 Requirements: KnightSat C&DH RequirementSourceVerification Source Document Test/Analysis Number S1.6-1Provide reliable, real-time access and control.
REAL-TIME SOFTWARE SYSTEMS DEVELOPMENT Instructor: Dr. Hany H. Ammar Dept. of Computer Science and Electrical Engineering, WVU.
Multiple Processor Systems. Multiprocessor Systems Continuous need for faster and powerful computers –shared memory model ( access nsec) –message passing.
1 System Models. 2 Outline Introduction Architectural models Fundamental models Guideline.
IMPROUVEMENT OF COMPUTER NETWORKS SECURITY BY USING FAULT TOLERANT CLUSTERS Prof. S ERB AUREL Ph. D. Prof. PATRICIU VICTOR-VALERIU Ph. D. Military Technical.
Computer Science Open Research Questions Adversary models –Define/Formalize adversary models Need to incorporate characteristics of new technologies and.
Cluster Reliability Project ISIS Vanderbilt University.
Architecting Web Services Unit – II – PART - III.
Source: George Colouris, Jean Dollimore, Tim Kinderberg & Gordon Blair (2012). Distributed Systems: Concepts & Design (5 th Ed.). Essex: Addison-Wesley.
© Oxford University Press 2011 DISTRIBUTED COMPUTING Sunita Mahajan Sunita Mahajan, Principal, Institute of Computer Science, MET League of Colleges, Mumbai.
Architectural Design lecture 10. Topics covered Architectural design decisions System organisation Control styles Reference architectures.
In-Vehicle Communication SAN Group RTS Regular Meeting Presentation December 2008.
Advanced Computer Networks Topic 2: Characterization of Distributed Systems.
B3AS Joseph Lewthwaite 1 Dec, 2005 ARL Knowledge Fusion COE Program.
CONTI'20041 Event Management in Distributed Control Systems Gheorghe Sebestyen Technical University of Cluj-Napoca Computers Department.
SOIS Plenary Session Spring 2015 Meeting Chris Taylor ESA.
Computers Operating System Essentials. Operating Systems PROGRAM HARDWARE OPERATING SYSTEM.
TTP and FlexRay. Time Triggered Protocols Global time by fault tolerant clock synchronisation Exact time point of a certain message is known (determinism)
REAL-TIME SOFTWARE SYSTEMS DEVELOPMENT Instructor: Dr. Hany H. Ammar Dept. of Computer Science and Electrical Engineering, WVU.
DSL Distributed Systems Laboratory ATC 23 August Model Mission: Magnetospheric Multiscale (MMS) Mission Goal “To study the microphysics of three.
SOIS Workshop The SOIS Communication Service Architecture Chris Plummer.
RIU as related to SOIS EDS Glenn Rakow CCSDS SOIS Spring Meeting 2013.
70-293: MCSE Guide to Planning a Microsoft Windows Server 2003 Network, Enhanced Chapter 12: Planning and Implementing Server Availability and Scalability.
Introduction to Grids By: Fetahi Z. Wuhib [CSD2004-Team19]
Advantages of Time-Triggered Ethernet
1 Software Design Lecture What’s Design It’s a representation of something that is to be built. i.e. design  implementation.
Software Engineering Chapter: Computer Aided Software Engineering 1 Chapter : Computer Aided Software Engineering.
Tolerating Communication and Processor Failures in Distributed Real-Time Systems Hamoudi Kalla, Alain Girault and Yves Sorel Grenoble, November 13, 2003.
ESA UNCLASSIFIED – For Official Use SOIS EDS Interoperability F. Torelli CCSDS Spring Meeting /03/2015.
CSC 480 Software Engineering Lecture 17 Nov 4, 2002.
Data and Applications Security Developments and Directions Dr. Bhavani Thuraisingham The University of Texas at Dallas Lecture #25 Dependable Data Management.
1 CEG 2400 Fall 2012 Network Servers. 2 Network Servers Critical Network servers – Contain redundant components Power supplies Fans Memory CPU Hard Drives.
SOIS Services Version 3, with post 19 Jan 2016 Telecon mods.
Seminar On Rain Technology
Ensuring Reliable Networks Copyright © TTTech Computertechnik AG. All rights reserved.Page 1 Christian Fidi Product Manager Advantages cFS.
SEMINAR TOPIC ON “RAIN TECHNOLOGY”
70-293: MCSE Guide to Planning a Microsoft Windows Server 2003 Network, Enhanced Chapter 12: Planning and Implementing Server Availability and Scalability.
Deterministic Communication with SpaceWire
Architecting Web Services
Architecting Web Services
Distribution and components
Modeling and Simulation of TTEthernet
DETERMINISTIC ETHERNET FOR SCALABLE MODULAR AVIONICS
Cyber Physical Systems
Mark McKelvin EE249 Embedded System Design December 03, 2002
Design Yaodong Bi.
Time-Triggered Architecture
Design.
Presentation transcript:

Time Triggered Networks: use in space 2015 CCSDS spring SOIS Plenary 23 March 2015 Glenn Rakow/NASA-GSFC

Contents Critical Vehicle Networks (CVN) or Deterministic Networks (DN) what does CVN? Fault tolerant vs non-fault tolerant DN application space CVN (DN) BOF purpose? describing new subnetwork services? Or are this a fundamental different type of data link,,not a service that any protocol may implement (at least easily without extensive modification) Fault tolerant Time-Triggered (TT) protocols – How are they different than other non-fault tolerant? What do TT protocols enable –TTA

CVN, DN Critical Vehicle Network (CVN) – what are the characteristics of a CVN? Deterministic Network (DN) – different methods on how to do accomplish and each method has different fault tolerance (or none at all) i.e. software schedule for master/slave transactions over Mil-Bus i.e, Time Triggered Protocol – no bus master – global time is maintained by collection of nodes

CVN, DN Application Space Crew-rated missions and robotic mission have very different requirements for CVN or DN Crew-rated mission have safety critical requirements that require “fly- through” faults/failures (fail operate) that are met through availability computing (multiple hot computers operating synchronously) Robotic mission do not have safety critical requirements Does it make sense to use the same protocol for both crew-rated and robotic missions? Many advantages for robotic missions in composability (foster reuse and therefore decrease development and integration time)

CVN (DN) BOF purpose? Seems that impetus of BOF is to address the crew-rated requirements Or is it meant for robotics as well (very different requirements)? Protocols already exist that serve both mission spaces Is purpose of BOF to to add services to subnetwork layer to account for DN (different services depending upon fault tolerance requirements)?

Fault Tolerant Time-Triggered Protocols Different class of DN protocols used for safety critical CVN applications (e.g., Flexray, TTCAN, SAFEbus, TTP/C, TTE) Two that currently stand-out as top contenders Standardized under SAE AS6802 – time triggered Gigabit Ethernet – used on Orion AS6003 – Time triggered protocol slash C – used for control surfaces for some regional jets TT protocols implement special services to ensure fault masking, i.e., Schedule Dispatch – message dispatch static schedule generated off-line for each component Clock synchronization – no time master Start-up Service – synchronize components (cold start or integration/reintegration) distributed algorithm so as to not rely on any one component External synchronization – allows onboard comm. to integrate to external time source, i.e, GPS Clique detection and resolution – unintended scenario where subset of components synchronize within the subset but not over subset boundaries Membership – low level diagnosis that continually monitor system’s health state, which nodes are present due to permanent/transient failures Integrity – bus guardian to protect against babble idiot, and end-to-end arguments Availability – redundancy management of communication channels

What do Time-Triggered Protocols Enable Time Triggered Architectures (TTA) – provides a computing framework for dependable distributed embedded systems Real time application decomposed into cluster(s) of nodes and a fault tolerant global time base generated of know precision at each node Time based used to specify interfaces among nodes 2 phase design approach – architecture design and component (node) design Architecture design phase – interactions among distributed components (and interfaces) in value domain and time domain fully specified Component design phase – Implemented taking these interface specifications as constraints For human-rated missions, safety critical application Availability computing for “fly-through” faults For robotic missions, distributed computing, faster development and integration times, better FDIR Using fault tolerant TT protocols along with time-space partitioned OS and software components

TTP/C vs. TTE Have different performance domains TTGbE is higher performance and higher SWaP Orion (crew-rated domain) TTP/C is lower performance and lower SWaP Crew-rated domain Robotic domain Both should be able to exist without proper layering of architecture Other protocols like Milbus, SpaceWire and SpaceFiber can implement DN services but with lower fault tolerance