Enhancing the Monitoring of Real-Time Performance in Linux

There is a growing trend in applying Linux operating system in the domain of embeddedsystems. This is due to the important features that Linux benets from, such as beingopen source, its light weight compared to other major operating systems, its adaptabilityto dierent platforms, and its more stable performance speed. However, there are up-grades that still need to be done in order to use Linux for real-time purposes. A numberof dierent approaches have been suggested in order to improve Linux's performance inreal-time environment. Nevertheless, proposing a correct-by-construction system is verydicult in real-time environment, mainly due to the complexity and unpredictability ofthem. Thus, run-time monitoring can be a helpful approach in order to provide the userwith data regarding to the actual timing behavior of the system which can be used foranalysis and modication of it. In this thesis work, a design for run-time monitoringis suggested and implemented on a real-time scheduler module that assists Linux withreal-time tasks. Besides providing crucial data regarding the timing performance of thesystem, this monitor predicts violations of timing requirements based on the currenttrace of the system performance.

[1]  Ann Q. Gates,et al.  A taxonomy and catalog of runtime software-fault monitoring tools , 2004, IEEE Transactions on Software Engineering.

[2]  Donal Heffernan,et al.  A monitoring approach to facilitate run-time verification of software in deeply embedded systems , 2010 .

[3]  Y. Liao,et al.  A Specificational Approach to High Level Program Monitoring and Measuring , 1992, IEEE Trans. Software Eng..

[4]  F. Halsall,et al.  Performance monitoring and evaluation of large embedded systems , 1987, Softw. Eng. J..

[5]  Werner Klauser,et al.  The /proc file system , 1998 .

[6]  Ivan Cibrario Bertolotti,et al.  Experimental evaluation of the Linux RT Patch for real-time applications , 2009, 2009 IEEE Conference on Emerging Technologies & Factory Automation.

[7]  Hideyuki Tokuda,et al.  An integrated time-driven scheduler for the ARTS kernel , 1989, Eighth Annual International Phoenix Conference on Computers and Communications. 1989 Conference Proceedings.

[8]  M. F.,et al.  Bibliography , 1985, Experimental Gerontology.

[9]  Shinpei Kato,et al.  ExSched: An External CPU Scheduler Framework for Real-Time Systems , 2012, 2012 IEEE International Conference on Embedded and Real-Time Computing Systems and Applications.

[10]  Mikael Sjödin,et al.  Run-Time Monitoring of Timing Constraints: A Survey of Methods and Tools , 2013, ICSEA 2013.

[11]  Teofilo F. Gonzalez,et al.  Performance data collection using a hybrid approach , 2005, ESEC/FSE-13.

[12]  Jens-Henrik Lindskov,et al.  Real-Time Linux in an Embedded Environment A Port and Evaluation of RTAI on the CRIS Architecture , 2003 .

[13]  George Lima,et al.  Evaluation of interrupt handling timeliness in real-time Linux operating systems , 2008, OPSR.

[14]  Guoyin Zhang,et al.  Study and Comparison of the RTHAL-Based and ADEOS-Based RTAI Real-time Solutions for Linux , 2006, First International Multi-Symposiums on Computer and Computational Sciences (IMSCCS'06).

[15]  Björn B. Brandenburg,et al.  LITMUS RT : A Status Report ∗ , 2007 .

[16]  Campus de Viesques,et al.  Worst-Case Utilization Bound for EDF Scheduling on Real-Time Multiprocessor Systems , 2000 .

[17]  Hideyuki Tokuda,et al.  A real-time tool set for the ARTS kernel , 1988, Proceedings. Real-Time Systems Symposium.

[18]  G. Lipari,et al.  Linux and Real-Time : Current Approaches and Future Opportunities , 2006 .

[19]  Dieter Haban,et al.  A Hybrid Monitor for Behavior and Performance Analysis of Distributed Systems , 1990, IEEE Trans. Software Eng..

[20]  Mikael Sjödin,et al.  Monitoring capabilities of schedulers in model-driven development of real-time systems , 2012, Proceedings of 2012 IEEE 17th International Conference on Emerging Technologies & Factory Automation (ETFA 2012).

[21]  Hennadiy Leontyev,et al.  Generalized tardiness bounds for global multiprocessor scheduling , 2007, 28th IEEE International Real-Time Systems Symposium (RTSS 2007).

[22]  Helmut Rzehak,et al.  STEPS TOWARDS A FULLY PREEMPTABLE LINUX KERNEL , 2003 .

[23]  Robert Love Lowering latency in Linux: introducing a preemptible kernel , 2002 .

[24]  D. Niehaus Hrtimers and Beyond : Transforming the Linux Time Subsystems , 2009 .

[25]  Jeffrey J. P. Tsai,et al.  A noninvasive architecture to monitor real-time distributed systems , 1990, Computer.

[26]  Robert Love,et al.  Linux Kernel Development , 2003 .