INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 1 Steffen Stein, TU Braunschweig, 2009 A Timing-Aware Update Mechanism for Networked Real-Time Systems.

Präsentation zum Thema: "INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 1 Steffen Stein, TU Braunschweig, 2009 A Timing-Aware Update Mechanism for Networked Real-Time Systems."—  Präsentation transkript:

1 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 1 Steffen Stein, TU Braunschweig, 2009 A Timing-Aware Update Mechanism for Networked Real-Time Systems Steffen Stein, Moritz Neukirchner, Harald Schrom, Rolf Ernst

2 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 2 Steffen Stein, TU Braunschweig, 2009 Outline Motivation, Problem Statement Approach Challenges Framework Architecture, Update Protocol Demonstrator description Demonstration Conclusion

3 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 3 Steffen Stein, TU Braunschweig, 2009 Motivation Embedded Systems are Software-Intense Many variants in e.g. automotive systems In-Field Updates are becoming common How do we ensure correctness? Lab Testing for all variants –Becomes infeasible with number of variants –Each update potentially adds a variant Our proposition –Trustable update procedure –Ensure correctness of individuum

4 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 4 Steffen Stein, TU Braunschweig, 2009 RTE Platform CPU contracting interface application 1 application n requestassert contract contracting interface requestassert contract CPU Paradigm: Contracting Separation of Platform and Application –Applications request resources and assert meeting of contraints –Platform guarantees performance –Contracts Application can be accepted if sufficient platform resources exist Runtime Environment as broker Here: Focus on Performance Aspects

5 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 5 Steffen Stein, TU Braunschweig, 2009 General Idea of Update Flow Heterogeneous Networked Embedded System Model-Based Feasibility Check engine Contract interface Contract Repository Watchdogs Applications Annotations WCET Comm. Partners Constraints … arg = arg; err = CANOS_Init(); if (err != CANOS_NO_ERR) { CanBusFreeLst = NULL; CanBusUsedLst = NULL; result = err; … Software Update Monitoring Adaptation

6 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 6 Steffen Stein, TU Braunschweig, 2009 Challenges in Distributed Performance Analysis Distributed Algorithm for Performance Analysis [Isola2006] –Exchange Event Stream Data –Local Analysis as soon as new data is available Compare distributed and centralized Algorithm –Convergence Properties (submitted to EMSOFT) –Quality of results does one of the algorithms yield tighter bounds? Consider runtime of Distributed Performance Analysis –Busy-Window Analysis as key component –Global iteration Convergence Detection

7 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 7 Steffen Stein, TU Braunschweig, 2009 Challenges in Update Management Develop an Architecture to cope with Updates –Strict Separation of Model and Execution Domain –Communication over Contract Repository Description of Software (Contracts) –Contents –Representation Development of Update Protocol –Interaction between Architecture Components Synchronization –Exploit task graph information –Barrier Synchronization

8 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 8 Steffen Stein, TU Braunschweig, 2009 Outline Motivation, Problem Statement Approach Challenges Framework Architecture, Update Protocol Demonstrator description Demonstration Conclusion

9 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 9 Steffen Stein, TU Braunschweig, 2009 General Architecture Update Controller Task Management Model Management Contract Repository Model domainExecution domain Application code & model model code contract Contract interface

10 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 10 Steffen Stein, TU Braunschweig, 2009 Model Domain Update Controller Contract Repository Model domain model contract Contract interface Model Optimization Model Analysis model metrics Contract Negotiation

11 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 11 Steffen Stein, TU Braunschweig, 2009 Update Protocol 3. Model Contract Enforcement 1. Annos 2. Test Annos 8. OK 6. Evaluation 9. Code 12. Configuration 13. Done Analysis Framework Contract Negotiation Update Controller Contract Repository Tasks 5. Analysis 11. Parameters 4. Application wide synchronisation 7. Application wide synchronisation 10. Application wide synchronisation

12 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 12 Steffen Stein, TU Braunschweig, 2009 Outline Motivation, Problem Statement Approach Challenges Framework Architecture, Update Protocol Demonstrator description Demonstration Conclusion

13 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 13 Steffen Stein, TU Braunschweig, 2009 RTE Implementation Work Plain C Based on RT-Kernel uC/OS-II Flexible Communication Infrastructure –Supports task migration –API inspired by AUTOSAR Task and Contract Management as outlined before –Contract Management –Task Migration Distributed Analysis Engine –Basic concepts (SPP Analysis, PJD Models)

14 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 14 Steffen Stein, TU Braunschweig, 2009 Physical Demonstrator Setup Stepper motors Angle sensor Track w/ photosensors IR illumination

15 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 15 Steffen Stein, TU Braunschweig, 2009 Demonstrator Application Setup Insertion of second Application breaks timing constraints of first Framework detects this Second Application is denied Left Controller D1D1 C Sensor C Comm C Ctrl. T1T1 T2T2 T3T3 Right Controller CAN-Bus Analysis Core 1 Analysis Core 2

16 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 16 Steffen Stein, TU Braunschweig, 2009 Demo Working Demo – see Demonstrator

17 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 17 Steffen Stein, TU Braunschweig, 2009 Performance Data Update verification, acceptance and integration in fractions of a second Most of the time spent for CAN communication (~ 90%) Memory footprint of the RTE (unoptimized) –30kB für uC/OS –30kB Analysis Engine –170kB Contracting and Runtime Framework –100kB Data (scalable) –Stacks not considered

18 INSTITUT FÜR DATENTECHNIK UND KOMMUNIKATIONS- NETZE 18 Steffen Stein, TU Braunschweig, 2009 Conclusion Distributed Performance Analysis Contracting Framework Task migration Integrated in a runtime Framework Enables performance-safe updates Ongoing work –Online Optimization –Distributed Performance Analysis –Online Monitoring