Ying Zhang&Krishnendu Chakrabarty Presenter Kasım Sert

Slides:



Advertisements
Similar presentations
EE5900 Advanced Embedded System For Smart Infrastructure
Advertisements

Time Demand Analysis.
Overload Scheduling in Real-Time Systems
CprE 458/558: Real-Time Systems (G. Manimaran)1 CprE 458/558: Real-Time Systems Fault-Tolerant Scheduling Techniques.
CprE 458/558: Real-Time Systems (G. Manimaran)1 CprE 458/558: Real-Time Systems Resource Access Control Protocols.
Courseware Scheduling of Distributed Real-Time Systems Jan Madsen Informatics and Mathematical Modelling Technical University of Denmark Richard Petersens.
Real Time Scheduling Terminologies define Fixed Priority Scheduler
Mehdi Kargahi School of ECE University of Tehran
Investigating the Effect of Voltage- Switching on Low-Energy Task Scheduling in Hard Real-Time Systems Paper review Presented by Chung-Fu Kao.
Present by Chen, Ting-Wei Adaptive Task Checkpointing and Replication: Toward Efficient Fault-Tolerant Grids Maria Chtepen, Filip H.A. Claeys, Bart Dhoedt,
7. Fault Tolerance Through Dynamic or Standby Redundancy 7.5 Forward Recovery Systems Upon the detection of a failure, the system discards the current.
By Group: Ghassan Abdo Rayyashi Anas to’meh Supervised by Dr. Lo’ai Tawalbeh.
Scheduling policies for real- time embedded systems.
Real-Time Scheduling CS4730 Fall 2010 Dr. José M. Garrido Department of Computer Science and Information Systems Kennesaw State University.
Survey of Real Time Databases Telvis Calhoun CSc 6710.
Hard Real-Time Scheduling for Low- Energy Using Stochastic Data and DVS Processors Flavius Gruian Department of Computer Science, Lund University Box 118.
CprE 458/558: Real-Time Systems (G. Manimaran)1 CprE 458/558: Real-Time Systems RMS and EDF Schedulers.
KUKUM Real Time System 1/21 Module 2 Real Time System Scheduling Lecture 05.
Introduction to Embedded Systems Rabie A. Ramadan 5.
CSCI1600: Embedded and Real Time Software Lecture 23: Real Time Scheduling I Steven Reiss, Fall 2015.
Introduction to Real-Time Systems
A Fault-Tolerant Scheduling Algorithm for Real-Time Periodic Tasks with Possible Software Faults Ching-Chih Han, Kang G. Shin, and Jian Wu.
Determining Optimal Processor Speeds for Periodic Real-Time Tasks with Different Power Characteristics H. Aydın, R. Melhem, D. Mossé, P.M. Alvarez University.
Academic Year 2014 Spring Academic Year 2014 Spring.
Fault-Tolerant Rate- Monotonic Scheduling Sunondo Ghosh, Rami Melhem, Daniel Mosse and Joydeep Sen Sarma.
Real-Time Operating Systems RTOS For Embedded systems.
Embedded System Scheduling
REAL-TIME OPERATING SYSTEMS
Chapter 19: Real-Time Systems
Advanced Operating Systems CIS 720
Prepared by Ertuğrul Kuzan
Prabhat Kumar Saraswat Paul Pop Jan Madsen
Wayne Wolf Dept. of EE Princeton University
EEE 6494 Embedded Systems Design
Presented by: Suresh Vadlakonda Ramanjaneya Gupta Pasumarthi
Imprecise Computation September 7, 2006
Lecture 24: Process Scheduling Examples and for Real-time Systems
Period Optimization for Hard Real-time Distributed Automotive Systems
Data Replication in the Quality Space
Department of Computer Science & Engineering
Clock-driven Static scheduling
Integer Programming (정수계획법)
Clock-driven Static scheduling
Real Time Scheduling Mrs. K.M. Sanghavi.
Clock-driven Static scheduling
Elastic Task Model For Adaptive Rate Control
CSCI1600: Embedded and Real Time Software
CSCI1600: Embedded and Real Time Software
Networked Real-Time Systems: Routing and Scheduling
Richard Anderson Lecture 6 Greedy Algorithms
Clock-driven Static scheduling
Chapter 19: Real-Time Systems
Richard Anderson Autumn 2016 Lecture 7
CSCI1600: Embedded and Real Time Software
Integer Programming (정수계획법)
Processes and operating systems
Richard Anderson Lecture 7 Greedy Algorithms
Chapter 10 Multiprocessor and Real-Time Scheduling
NET 424: REAL-TIME SYSTEMS (Practical Part)
NET 424: REAL-TIME SYSTEMS (Practical Part)
Richard Anderson Winter 2019 Lecture 7
M. Kezunovic (P.I.) S. S. Luo D. Ristanovic Texas A&M University
Richard Anderson Autumn 2015 Lecture 7
The End Of The Line For Static Cyclic Scheduling?
Ch 4. Periodic Task Scheduling
Operating Systems Concepts
Communication Driven Remapping of Processing Element (PE) in Fault-tolerant NoC-based MPSoCs Chia-Ling Chen, Yen-Hao Chen and TingTing Hwang Department.
Anand Bhat*, Soheil Samii†, Raj Rajkumar* *Carnegie Mellon University
Real-Time Scheduling David Ferry CSCI 3500 – Operating Systems
Richard Anderson Autumn 2019 Lecture 7
Presentation transcript:

Fault Recovery Based on Checkpointing for Hard Real-Time Embedded Systems Ying Zhang&Krishnendu Chakrabarty Presenter Kasım Sert Spring 2007, Boğaziçi University

Introduction The checkpointing of system state and restoring it in the case of a system fault is one method if goal is the creation of a robust, fault-tolerant system. Safety-critical embedded systems often necessitate fault-tolerant computing techniques because of their harsh operating environment. The correctness of these systems depends not only on the result of computation but also on the time that result is produced. Fault tolerance is achieved by; Online fault detection Checkpointing & Rollback

Introduction Why fault tolerance is needed ? Checkpointing; (+)Increases task execution time and may cause missing deadline which completes on time without checkpointing. (-)If checkpointing is chosen carefully, it may cause on time termination although faults occur. Why fault tolerance is needed ? Rapid increase in processor speeds. Dependence of correctness to the timely completion of underlying tasks. Lower processor voltages => Lower noise margins.

Checkpointing in real-time systems Offline-Checkpointing schemes Checkpointing interval is known before task execution. Online-Checkpointing Checkpointing interval can be adapted to fault occurrences. However, interval rates are generally considered as probabilistic i.e. Poisson process, equidistant..

Off-line Checkpointing Analysis Γ = { τ1, τ2, …, τn} of n periodic real-time tasks τi = (Ti, Di, Ei). Ti is the period of τi, Di is its deadline (Di ≤ Ti), and Ei is the execution time Proposed approaches; 1)To tolerate k faults for each job, termed as job-oriented fault tolerance; 2)To tolerate k faults within a hyper-period (defined as the least common multiple of all the task periods, termed as hyper-period-oriented fault tolerance.

Job-Oriented Fault Tolerance 1-Compute where Th is period and Eh is execution time. 2-The iteration is terminated either when or In the former case, τi is schedulable; in the latter case, τi is not schedulable. Under faulty conditions, the additional time due to checkpointing and recovery should be incorporated.

Hyperperiod Oriented Fault Tolerance Start from the highest priority task and calculate the minimum number of checkpoints mi to make it schedulable. Calculate the response time Ri. If Ri <= Di move to the next task Else continue reducing Ri  To reduce Ri add more checkpoints.

Comment on the Algorithm There are two key issues which are not addressed in the algorithm. Checkpoints are added to highest priority task. However they may belong to the earlier iteration tasks, in this situation calculation after that task may not be valid. Adding more checkpoints may not always improve overall execution time.

Solving the Deficiencies of Algorithms To solve the first problem addressed above; all lower-priority tasks need to be re-examined. To solve the second one; Analysis of a bound based on checkpointing tradeoffs. Analysis of a bound based on timing constraints.

Proposed Solution Algorithm

Experimental Results The performance of proposed fault tolerant schemes (JFT,HFT) are compared with rate-monothonic (RM) schemes. It is assumed that RM simply re-executes a job a fault occurs.

Experimental Results-JFT vs. RM Re-execution takes extra time and total task utilization falls under 1. Thus, in the presence of faults RM is not schedulable.

Experimental Results-HFT vs. RM

Thank You for Listening Any Questions?