event triggered tasks
play

event-triggered tasks Martijn van den Heuvel Reinder J. Bril - PowerPoint PPT Presentation

ETFA 2013 Limited preemptive scheduling of mixed time-triggered and event-triggered tasks Martijn van den Heuvel Reinder J. Bril Xiaodi Zhang Syed Md Jakaria Abdullah Damir Isovic Complexity of industrial applications? Efficient resource


  1. ETFA 2013 Limited preemptive scheduling of mixed time-triggered and event-triggered tasks Martijn van den Heuvel Reinder J. Bril Xiaodi Zhang Syed Md Jakaria Abdullah Damir Isovic

  2. Complexity of industrial applications? Efficient resource usage vs complex constraints / Mathematics and computer science 9-9-2013 PAGE 1

  3. Time-triggered task model • Job • Absolute deadline = end of interval I 0 I 1 I 2 I 4 I 0 I 3 ... 0 4 6 9 14 20 • Execute jobs in the slots captured in a scheduling table • Resolve complex task constraints offline / Mathematics and computer science 9-9-2013 PAGE 2

  4. Offline scheduling • Offline reservation of time-triggered tasks into intervals I 0 I 1 P 1 P 2 P 1 P 3 P 2 P 3 0 1 2 3 4 5 6 7 8 9 10 11 12 • Each interval is divided into • spare capacity and • reserved capacity length of the interval reserved capacity spare capacity Time-triggered tasks / Mathematics and computer science 9-9-2013 PAGE 3

  5. Time-triggered task model • Job • Absolute deadline = end of interval I 0 I 1 I 2 I 4 I 0 I 3 ... 0 4 6 9 14 20 • Execute jobs according to a scheduling table • Resolve complex task constraints offline What to do with the free capacity? Polling is expensive → efficient event handling? How to extend system functionality? / Mathematics and computer science 9-9-2013 PAGE 4

  6. Mixed time- and event-triggered Task model I 0 I 1 I 2 I 3 I 4 I 0 ... Time-triggered (TT): 0 4 6 9 14 20 Sporadic task ( S i ) Event-triggered (ET): A-periodic task ( A i ) • Use the free time slots for ET jobs: D i • Relative deadline = D i job • Absolute deadline = arrival time + D i Arrival time d i,k Preemptive EDF scheduling: based on absolute deadline / Mathematics and computer science 9-9-2013 PAGE 5

  7. How (fully preemptive) slotshifting works? [Isovic and Fohler 2009] / Mathematics and computer science 9-9-2013 PAGE 6

  8. How (fully preemptive) slotshifting works? [Isovic and Fohler 2009] / Mathematics and computer science 9-9-2013 PAGE 7

  9. How (fully preemptive) slotshifting works? [Isovic and Fohler 2009] / Mathematics and computer science 9-9-2013 PAGE 8

  10. Slotshifting versus static offline schedules • Advantages of slotshifting: • Online reclaiming of spare capacity • Same worst-case offline analysis as static schedules • Disadvantages of slotshifting: • Shifting of slots must allow for arbitrary preemptions • No support for allocating subsequent slots Non-preemptive shifts How to overcome these disadvantages? / Mathematics and computer science 9-9-2013 PAGE 9

  11. Slotshifting versus static offline schedules • Advantages of slotshifting: • Online reclaiming of spare capacity • Same worst-case offline analysis as static schedules • Disadvantages of slotshifting: • Shifting of slots must allow for arbitrary preemptions • No support for allocating subsequent slots Non-preemptive shifts Restricted shifting of TT tasks! See paper… / Mathematics and computer science 9-9-2013 PAGE 10

  12. Problem description • What already exists? • Slotshifting of preemptive tasks [Isovic and Fohler 2009] • What is new? • Extended task analysis for non-preemptive execution Local system Global system Time-triggered Offline scheduling New offline analysis+ New online scheduling Event-triggered Standard (SRP +) EDF mechanism scheduling / Mathematics and computer science 9-9-2013 PAGE 11

  13. Problem description • What already exists? • Slotshifting of preemptive tasks [Isovic and Fohler 2009] • What is new? • Extended task analysis for non-preemptive execution Local system Global system Time-triggered Offline scheduling New offline analysis+ New online scheduling Event-triggered Standard (SRP +) EDF mechanism scheduling / Mathematics and computer science 9-9-2013 PAGE 12

  14. Problem: how to disable preemptions? Global system Local TT task system Local ET task system Globally Time-triggered tasks Event-triggered tasks Non-preemptive Preemption Locally Locally Non-preemptive Non-preemptive Currently: fully preemptive EDF of the global system Our contributions: 3 limited-preemptive extensions / Mathematics and computer science 9-9-2013 PAGE 13

  15. Problem: how to disable preemptions? Global system Local TT task system Local ET task system Globally Time-triggered tasks Event-triggered tasks Non-preemptive Preemption Locally Locally Non-preemptive Non-preemptive 1. Restricted shifting of TT tasks! See paper… / Mathematics and computer science 9-9-2013 PAGE 14

  16. Global non-preemptive execution of ET tasks Global system Local TT task system Local ET task system Globally Time-triggered tasks Non-preemptive Event-triggered tasks Locally Locally Non-preemptive Non-preemptive New offline analysis + Our main New online scheduling mechanism contribution: / Mathematics and computer science 9-9-2013 PAGE 15

  17. Problem: how to disable preemptions? Global system Local TT task system Local ET task system Globally Time-triggered tasks Event-triggered tasks Non-preemptive Preemption Locally Locally Non-preemptive Non-preemptive 2. Intermediate step: Incorporate EDF+SRP analysis of [Baruah, RTSS 2006] / Mathematics and computer science 9-9-2013 PAGE 16

  18. Offline analysis of ET tasks (Recap) [Isovic and Fohler 2009] / Mathematics and computer science 9-9-2013 PAGE 17

  19. Slotshifting: a fully preemptive system (Recap) Global system Local TT task system Local ET task system Admission? Time-triggered tasks Event-triggered tasks I 0 I 1 I 2 I 4 I 3 I 0 ... 0 4 6 9 14 20 / Mathematics and computer science 9-9-2013 PAGE 18

  20. Offline analysis of ET tasks (Recap) Schedulability test of sporadic tasks • Classical demand-based analysis • Supply based on static offline schedule Demand ≤ L [Baruah 2006] during an interval of length L Supply based on offline schedule Supply Demand ≤ during an interval of length L during an interval of length L / Mathematics and computer science 9-9-2013 PAGE 19

  21. Calculation of supply bound ( t 2 − t 1 ) − overlapping reserved capacities in d intervals L I 0 I 1 I 2 I 3 t 2 t 1 L I 0 I 1 I 2 I 3 t 2 t 1 L I 0 I 1 I 2 I 3 t 2 t 1 / Mathematics and computer science 9-9-2013 PAGE 20

  22. Calculation of supply bound L = ( t 2 − t 1 ) L I 0 I 1 I 2 I 3 t 2 t 1 L I 0 I 1 I 2 I 3 t 2 t 1 L I 0 I 1 I 2 I 3 t 2 t 1 / Mathematics and computer science 9-9-2013 PAGE 21

  23. Schedulability of ET tasks • A sporadic task set T s is can be scheduled under EDF together with an offline schedule, iff where − Demand bound function, dbf [Baruah, RTSS 2006] − New Supply bound function, sbf Calculation of sbf: ( t 2 − t 1 ) − overlapping reserved capacities in d intervals / Mathematics and computer science 9-9-2013 PAGE 22

  24. Disable local preemptions of ET tasks Global system Local TT task system Local ET task system Time-triggered task Event-triggered task Preemption Locally Locally Non-preemptive Non-preemptive Intermediate step: Incorporate EDF+SRP analysis of [Baruah, RTSS 2006] / Mathematics and computer science 9-9-2013 PAGE 23

  25. Scheduling rules for local ET system Rule • If an ET task starts executing its critical section − Case 1: Preemptions from any other ET tasks are disabled − Case 2: Preemptions from TT task with higher priority are allowed / Mathematics and computer science 9-9-2013 PAGE 24

  26. Non-preemptive sections of local ET tasks • Local blocking: Reserved capacity of interval Sporadic task Critical section of sporadic task Blocking I 0 I 1 S 1 S 2 rc ℓ / Mathematics and computer science 9-9-2013 PAGE 25

  27. Non-preemptive sections of local ET tasks • Schedulability test • Blocking function [Baruah, RTSS 2006] If h j denotes length of critical sections B( L )  maximum h j needed by a lower priority task / Mathematics and computer science 9-9-2013 PAGE 26

  28. Global non-preemptive execution of ET tasks Global system Local TT task system Local ET task system Globally Time-triggered tasks Non-preemptive Event-triggered tasks Locally Locally Non-preemptive Non-preemptive New offline analysis + Our main New online scheduling mechanism contribution: / Mathematics and computer science 9-9-2013 PAGE 27

  29. Scheduling rules for Global System • Rule 1 • Case 1: spare capacity > critical section length ET task executes globally non-preemptively • Case 2: spare capacity < critical section length ET task blocks it-self and other ET tasks / Mathematics and computer science 9-9-2013 PAGE 28

  30. Scheduling rules for Global System • Rule 1 • Case 1: spare capacity > critical section length ET task executes globally non-preemptively • Case 2: spare capacity < critical section length ET task blocks it-self and other ET tasks • Rule 2 • Case 1: another ET task wants to execute during self-blocking ET task is not allowed to execute • Case 2: a TT task wants to execute during self-blocking TT task is allowed to execute / Mathematics and computer science 9-9-2013 PAGE 29

  31. Global non-preemptive execution of ET tasks • Self-blocking mechanism Reserved self-blocking slot Reserved capacity of interval Sporadic task Global critical section Miss Miss I 0 I 1 I 2 I 3 S 1 rc ℓ 1 6 9 16 / Mathematics and computer science 9-9-2013 PAGE 30

Recommend


More recommend