1. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 1 Quality of Service for In-Home Digital Networks PROGRESS PROJECT EES.5653 Terminal QoS M.A. Albu

2. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 2 Contents Summary work Terminal QoS Collaboration with MRM project Number of context switches estimation method Future work

3. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 3 Summary work Literature survey on QoS work QoS overview and classification of QoS improvement techniques – internal report Number of context switches estimation method: –1 st approach: statistical approach –2 nd approach: min-max method –3 rd approach: average method

4. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 4 Terminal QoS QoS determined by resource management of the system in discussion Terminal resources under investigation: –CPU, –Memory, –Bus

5. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 5 Collaboration with MRM project Why MRM? –QoS related closely to resource management. –MRM is concerned with resources management aspects in the context of a terminal. –MRM provides opportunities for inspiration, validation of my work Aim MRM: –provide methods and means for an integrated approach to resource management in multi-resource systems. The integrated approach has to meet at least the following requirements: –The resource management infrastructure should be able to provide resource guarantees to the building blocks of application functionality. –Individual building blocks should be able to limit or prevent resource insufficiencies, by dealing with insufficient resources in a graceful and predictable way.

6. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 6 Collaboration with MRM project Rationale  This leads to the need for developing methods for estimating the necessary of resources for the building blocks of application but also for predicting resources necessary for the composed execution of these blocks. Why performance composition? - Just adding clock cycles of the involved components won’t do.  Method for the estimation of the number of context switches occurring during the execution of a streaming application. Current experimentation setting: –HW: Trimedia (TM 1300) incorporates a media processor for high-performance multimedia applications that deal with high-quality video and audio. –SW: TSSA

7. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 7 MRM project - TSSA TSSA – TriMedia Streaming Software Architecture Component1Component2Component 3 FP Q EP Q

8. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 8 Reasons for context switch occurrence Blocking. The execution of a task blocks because of the following reasons: o Communication with the PC host (ex: FRead) o Unfavorable status of the queues: - input full packets queue (IFPQ) is empty (no input) - output full packets queue (OFPQ) is full (task cannot output packets for the moment) - output empty packets queue (OEPQ) is empty (task cannot output packets for the moment) Preemption. The execution of a task is preempted by another task with a higher priority. Task execution end. The execution of a task with high priority has ended (no preemption or blocking) and the resources are allocated to another task.

9. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 9 NCS Estimation Method Properties of streaming applications executions Property1: Running streaming applications, after an initialization phase, adopt a pattern of execution that repeats after a specific interval of time (hyperperiod). The repetitive execution is caused by the differences in the components’ rates of production/consumption of full/empty packets. Execution consists of 3 phases: initialization, stable-state, finalization => by knowing the NCS occuring during initialization, finalization and during a hyperperiod of the steady-state, we obtain the total NCS Property2: When one of the components in the streaming chain is periodic, when other components depend on it in execution, their tasks will execute periodic. Initialization Stable-state Finalization hyperperiod

10. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 10 NCS Estimation Method Case study description FReadVDecVRendVO FP Q EP Q

11. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 11 NCS Estimation Method Steps 1 – 4 : initialization, finalization phases Step 1. Initialization phase: End phase: 169.261 ms Duration: 169.261 ms Step 2. Initialization phase: NCS_initializationPhase(FRead) = 26 NCS_initializationPhase (VDec) = 22 NCS_initializationPhase (VRendVO) = 6 Step 3. Finalization phase: Beginning phase: 4539.029 ms Duration: 1569.428 ms Step 4. Finalization phase: NCS_finalizationPhase(FRead) = 8 NCS_finalizationPhase(VDec) = 356 NCS_finalizationPhase(VRendVO) = 94

12. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 12 NCS Estimation Method Steps 5 – 6 : stable-state phase, production/consumption rates Step 5. Stable state: Beginning phase: 169.261 ms End phase: 4539.029 ms Duration:4369.768 ms Step 6. Identify for each component the full packets production rate (FPPR), the full packets consumption rate (FPCR), and the empty packets production rate (EPPR). Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 2.524 Vdec 70 4.6 17.9 4.5 2 VrendVO 80 16.3 16.3 0.056 32.6 - measurements of components rates and computation times in isolation.

13. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 13 NCS Estimation Method Steps 5 – 6 : stable-state phase, production/consumption rates Step 5. Stable state: Beginning phase: 169.261 ms End phase: 4539.029 ms Duration:4369.768 ms Step 6. Identify for each component the full packets production rate (FPPR), the full packets consumption rate (FPCR), and the empty packets production rate (EPPR). Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 2.524 Vdec 70 4.6 17.9 4.5 2 VrendVO 80 16.3 2*FPPR(VO) 16.3 0.056 32.6

14. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 14 NCS Estimation Method Steps 5 – 6 : stable-state phase, production/consumption rates Step 5. Stable state: Beginning phase: 169.261 ms End phase: 4539.029 ms Duration:4369.768 ms Step 6. Identify for each component the full packets production rate (FPPR), the full packets consumption rate (FPCR), and the empty packets production rate (EPPR). Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 2.524 Vdec 70 4.6 17.9 4.5 2 VrendVO 80 16.3 2*FPPR(VO) 2*FPPR(VO) 16.3 0.056 32.6

15. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 15 NCS Estimation Method Steps 5 – 6 : stable-state phase, production/consumption rates Step 5. Stable state: Beginning phase: 169.261 ms End phase: 4539.029 ms Duration:4369.768 ms Step 6. Identify for each component the full packets production rate (FPPR), the full packets consumption rate (FPCR), and the empty packets production rate (EPPR). Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 2.524 Vdec 70 4.6 17.9 4*FPPR(VDec) 4.5 2 VrendVO 80 16.3 2*FPPR(VO) 2*FPPR(VO) 16.3 0.056 32.6

16. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 16 NCS Estimation Method Steps 5 – 6 : stable-state phase, production/consumption rates Step 5. Stable state: Beginning phase: 169.261 ms End phase: 4539.029 ms Duration:4369.768 ms Step 6. Identify for each component the full packets production rate (FPPR), the full packets consumption rate (FPCR), and the empty packets production rate (EPPR). Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 - - 2.524 - Vdec 70 4.6 17.9 4*FPPR(VDec) 4.5 2 VrendVO 80 16.3 2*FPPR(VO) 2*FPPR(VO) 16.3 0.056 32.6

17. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 17 NCS Estimation Method Step 7 : dependencies between components Step 7 a -Determine the dependencies in the execution of the components by taking into consideration FPPR, FPCR and EPPR for each component. b - Determining the dependencies in the execution of the components, leads to determining the period (T(Ti)) of each task Ti on which the components C i are mapped. VDec: a - FPPR (VDec) > FPCR (VRendVO) (> = rate higher) => OFPQ (VDec) at stable state is full => OEPQ (VDec) is empty => VDec depends on VRendVO to produce 1 EP so that VDec can produce 1 FP => FPPR (VDec) := EPPR (VRendVO) = 2 * FPPR(VrendVO) = 2 * T(VrendVO) = 32.6 ms. => b - Since T(Vdec) = FPPR(VDec) => T(Vdec) = 32.6 ms

18. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 18 NCS Estimation Method Step 7 : dependencies between components Step 7 a -Determine the dependencies in the execution of the components by taking into consideration FPPR, FPCR and EPPR for each component. b - Determining the dependencies in the execution of the components, leads to determining the period (T(Ti)) of each task Ti on which the components C i are mapped. FRead: a – FPPR (FRead) > FPCR (VDec) (> = rate higher) => OFPQ (FRead) at stable state is full => OEPQ (FRead) is empty => FRead depends on VDec to produce 1 EP so that FRead can produce 1 FP => FPPR (FRead) := EPPR (VDec) = 4 * FPPR(VDec) = 4 * T(VDec) = 4*2*T(VrendVO) :=130.4 ms => b - Since T(FRead) = FPPR(FRead) => T(FRead) = 130.4 ms

19. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 19 NCS Estimation Method Step 7 : dependencies between components Step 7 Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 - - 2.524 - Vdec 70 4.6 17.9 4*FPPR(VDec) 4.5 2 VrendVO 80 16.3 2*FPPR(VO) 2*FPPR(VO) 16.3 0.056 32.6

20. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 20 NCS Estimation Method Step 7 : dependencies between components Step 7 Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 2.2 - - 2.524 - Vdec 70 4.6 17.9 130.4 4.5 2 VrendVO 80 16.3 32.6 32.6 16.3 0.056 32.6

21. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 21 NCS Estimation Method Step 7 : dependencies between components Step 7 Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 130.4 - - 2.524 - Vdec 70 32.6 17.9 130.4 4.5 2 VrendVO 80 16.3 32.6 32.6 16.3 0.056 32.6

22. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 22 NCS Estimation Method Step 7 : dependencies between components Step 7 Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 130.4 - - 130.4 2.524 - Vdec 70 32.6 17.9 130.4 32.6 4.5 2 VrendVO 80 16.3 32.6 32.6 16.3 0.056 32.6

23. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 23 NCS Estimation Method Step 8 : hyperperiod length, number of hyperperiods Step 8. Identify hyperperiod length. CIS (component index set) = the set of natural numbers that serve as indexes for components in a streaming chain. The indexes of components will be equal with the indexes of the tasks on which the components are mapped at execution. HL = max T(T i ) = 8 * T(VRendVO) = 130.4 ms i  CIS Duration stable phase = 4369.768 ms => average number of hyperperiods during stable phase: HN =  Duration_stableStatePhase/HL  =  4369.768/HL  =  4369.768/130.4  = 34

24. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 24 NCS Estimation Method Step 9 : NCS due to blocking Step 9. Determine the NCS due to blocking. FRead: FRead blocks 4 times for each packet that it delivers due to communication with the PC host and has its period equal with the hyperperiod (because it only gets to deliveres 1 full packet during the hyperperiod after which it blocks until the next hyperperiod) => NCS_blocking(FRead) =  HL / T(Ti)  + NCS_inherentBlocking(FRead) =1 + 4 = 5;

25. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 25 NCS Estimation Method Step 9 : NCS due to blocking Step 9. VDec: VDec delivers 1 full packet every time it is activated after which it is blocked. VDec is activated periodically and that its period fits 4 time during the hyperperiod => VDec is activated 4 times during the hyperperiod => VDec is blocked 4 times during the hyperperiod. NCS_blocking(VDec) =  HL / T(Ti)  + NCS_inherentBlocking(VDec) =  HL / T(VDEc)  + 0 =  8*T(VRendVO) / 2*T(VRendVO)  = 4; VRendVO: VRendVO is the component that does not depend on any other component in its execution, and has no inherent blockings => it does not block. NCS_blocking(VRendVO) = 0;

26. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 26 NCS Estimation Method Step 10 : NCS due to normal execution end. Step 10. Determine the NCS due to normal execution. Applies only to components that are not preempted and do not depend on any other component in its execution, thus do not get blocked. => Applies only to VRendVO: NCS_normalExecutionEnd(VRendVO) =  HL / T(VRendVO)  =  8*T(VRendVO) / T(VRendVO)  = 8

27. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 27 NCS Estimation Method Step 11 : Task activation time (AT) Step 11. For each component by considering the dependencies dictated by the rates of production/consumption packets, calculate the first AT in the hyperperiod. In general: For each component Ci mapped on task Ti: 1. If  j  CIS | Ci dependent on Cj: If P(Ti) < P(Tj): AT(Ti) = AT(Tj) + (N-1)* FPPR(Tj) + CT(Tj) if Ti depends on Tj to release N FP. (N-1)* EPPR(Tj) + CT(Tj) if Ti depends on Tj to release N EP. TiTi TjTj CEPT(CFPT) EPPR(FPPR) CT-CEPT (CT-CFPT) CT-CEPT (CT-CFPT) N=4 CT

28. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 28 NCS Estimation Method Step 11 : Task activation time (AT) If P(Ti) > P(Tj): AT(Ti) = AT(Tj) + CFPT(Tj) + (N-1)* FPPR(Tj) if Ti depends on Tj to release N FP. CEPT(Tj) + (N-1)* EPPR(Tj) if Ti depends on Tj to release N EP. TiTi TjTj CEPT(CFPT) EPPR(FPPR) N=4 CT

29. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 29 NCS Estimation Method Step 11 : Task activation time (AT) For the current case study: FRead: FRead dependent on Vdec to release 1 EP => P(FRead) > P(VDec) => AT(FRead) = AT(VDec) + (1-1)*EPPR(Vdec) + CEPT (VDec) = AT(VDec) + CEPT(VDec) = AT(VDec) + 2ms. Vdec: Vdec dependent on VRendVO to release 1 EP => P(VRendVO) > P(VDec) => AT(VDec) = AT(VRendVO) + (1-1)*EPPR(VRendVO) + CT (VRendVO) = AT(VRendVO) + CT (VRendVO) = AT(VRendVO) + 0.056 ms. = 0.056 ms First AT(VRendVO) = 0 relative to the beginning of the hyperperiod. => AT(FRead) = 0.056 ms +2 ms = 2.056 ms

30. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 30 NCS Estimation Method Step 11 : Task activation time (AT) Step 11 Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 130.4 - - 130.4 2.524 - Vdec 70 32.6 17.9 130.4 32.6 4.5 2 VrendVO 80 16.3 32.6 32.6 16.3 0.056 32.6

31. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 31 NCS Estimation Method Step 11 : Task activation time (AT) Step 11 Priority FPPR FPCR EPPR T AT CT CEPT (ms) FRead 90 130.4 - - 130.4 2.056 2.524 - Vdec 70 32.6 17.9 130.4 32.6 0.056 4.5 2 VrendVO 80 16.3 32.6 32.6 16.3 0 0.056 32.6

32. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 32 NCS Estimation Method Task response time, NCS due to preemptions P(T i ) > P(T j )  AT(Ti)  (AT(Tj), AT(Tj) + CT(Tj)) => Ti preempts Tj. TiTi TjTj CT(Tj) AT(Tj) R 0 (Tj) = CT(Tj)

33. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 33 NCS Estimation Method Task response time, NCS due to preemptions P(T i ) > P(T j )  AT(Ti)  (AT(Tj), AT(Tj) + CT(Tj)) => Ti preempts Tj. TiTi TjTj AT(Tj)AT(Ti) R1(Tj) = Ro(Tj)+CT(Ti) CT(Ti)

34. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 34 NCS Estimation Method Task response time, NCS due to preemptions P(T i ) > P(T j )  AT(Ti)  (AT(Tj), AT(Tj) + CT(Tj)) => Ti preempts Tj. TiTi TjTj AT(Tj)AT(Ti) R1(Tj) = Ro(Tj)+2*CT(Ti) T(Ti) CT(Ti)

35. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 35 NCS Estimation Method Task response time, NCS due to preemptions P(T i ) > P(T j )  AT(Ti)  (AT(Tj), AT(Tj) + CT(Tj)) => Ti preempts Tj. NCS_preemption(Tj) =  Ro (Tj)/T(Ti)  R1(Tj) = NCS_preemption(Tj)*CT(Ti) TiTi TjTj AT(Tj)AT(Ti) R1(Tj) = Ro(Tj)+3*CT(Ti) T(Ti) CT(Ti) AT(Ti)

36. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 36 NCS Estimation Method Step 12:Task response time, NCS due to preemptions Step 12 Calculate NCS_preemption for all components: In general: R n (T i ) = R n-1 (T i ) +   R n-1 (T i )/T(Tj)  * CT(Tj), j  {k  CIS | P(T k ) > P(T i )  AT(Tk)  (AT(Ti), AT(Ti) + CT(Ti))} where Ro – initial response time, Ro( T i ) = CT( T i ) From here, the total number of context switches due to preemptions will be: NCS_preemption(Ti) =   R n-1 (Ti)/Tj  j  {k  CIS | P(T k ) > P(T i )  AT(Tk)  (AT(Ti), AT(Ti) + CT(Ti))}

37. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 37 NCS Estimation Method Step 12:Task response time, NCS due to preemptions Step 12 For the current case study: FRead: FRead has he highest priority assigned => never preempted. => NCS(FRead)_preemption = 0 VDec: P(VRendVO) > P(VDec)  AT(VRendVO), AT(VRendVO)  (AT(VDec), AT(VDec) + CT(VDec)) => VrendVO does not preempt VDec. P(FRead) > P(VDec)  AT(FRead), AT(FRead)  (AT(VDec), AT(VDec) + CT(VDec)) => FRead preempts VDec NCS_preemption (VDec) = 5 VRendVO: P(FRead) > P(VRendVO)  AT(FRead), AT(FRead)  (AT(VRendVO), AT(VRendVO) + CT(VRendVO)) => FRead does not preempt VRendVO. => NCS_preemption(VRendVO)=0

38. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 38 NCS Estimation Method Step 12:Task response time, NCS due to preemptions Step 12 For the current case study: FRead: FRead has he highest priority assigned => never preempted. => NCS(FRead)_preemption = 0 FRead - hybrid task that due to dependencies in execution activated every 130.4 ms. After activation it runs blocks 4 times /packet due to the communication with the PC host. The activation and blocking times after the first activation are the following: AT(FRead) 1 :CT1 = 0.121 ms, BT1 = 0.577 ms AT(FRead) 2 :CT2= 0.152 ms, BT2 = 0.499 ms AT(FRead) 3 :CT3 = 0.081 ms, BT3 = 0.126 ms AT(FRead) 4 :CT4 = 0.09 ms, BT4 = 0.774 ms AT(FRead) 5 :CT5 = 0.104 ms,  CT(FRead) =  CTi +  BTi = 2.524 ms

39. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 39 NCS Estimation Method Step 12:Task response time, NCS due to preemptions VDec: P(VRendVO) > P(VDec)  AT(VRendVO), AT(VRendVO)  (AT(VDec), AT(VDec) + CT(VDec)) => VrendVO does not preempt VDec. FRead - hybrid task => each of its 5 “internal/inherent” activations per period due to the communication with the PC host will be treated as independent tasks: FRead1, FRead2, …, FRead5. The 5 “tasks” have the same period – 130.4 ms = 8*T(VRendVO) and have equal priorities with the priority assigned to FRead. P(FRead) > P(VDec)  AT(FRead), AT(FRead)  (AT(VDec), AT(VDec) + CT(VDec)) => FRead preempts VDec

40. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 40 NCS Estimation Method Step 12:Task response time, NCS due to preemptions Ro = CT(VDec), Ro – initial response time of VDec NCS_preemption(VDec) 1 =  Ro (VDec)/T(FRead1)  =  4.5 / 130.4  = 1 R1 = Ro +  Ro (VDec)/T(FRead1)  * CT(FRead1) = 4.5 + 1* 0.121 = 4.621 ms NCS_preemption(VDec) 2 =  R1 (VDec)/T(FRead2)  =  4.621 / 130.4  = 1 R2 = R1 +  R1 (VDec)/T(FRead2)  * CT(FRead2) = 4.773 NCS_preemption(VDec) 3 =  R2 (VDec)/T(FRead3)  =  4.773 / 130.4  = 1 R3 = R2 +  R2 (VDec)/T(FRead3)  * CT(FRead3) = 4.854 ms NCS_preemption(VDec) 4 =  R3 (VDec)/T(FRead4)  =  4.854 / 130.4  = 1 R4 = R3 +  R3 (VDec)/T(FRead4)  * CT(FRead4) = 4.944 ms NCS_preemption(VDec) 5 =  R4 (VDec)/T(FRead5)  =  4.944 / 130.4  = 1  the 5 tasks that compose the FRead hybrid task preempt VDec each 1 time =>  NCS_preemption (VDec) = 5

41. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 41 NCS Estimation Method Step 12:Task response time, NCS due to preemptions VRendVO: P(FRead) > P(VRendVO)  AT(FRead), AT(FRead)  (AT(VRendVO), AT(VRendVO) + CT(VRendVO)) => FRead does not preempt VRendVO. => NCS_preemption(VRendVO)=0

42. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 42 NCS Estimation Method Step 13: Total NCS/hyperperiod Step 13 Determine NCS_total for each of the components involved: For each hyperperiod: NCS_total(Ci) = NCS_blocking(Ci) + NCS_preemption(Ci) + NCS_normalExecutionEnd(Ci) => Total NCS/hyperperiod: NCS_hyperperiod (FRead) = NCS_blocking(FRead) + NCS_preemption(FRead) = 5 + 0=5 NCS_hyperperiod (VDec) = NCS_blocking (VDec) + NCS_preemption (VDec) = 5 + 4 = 9 NCS_hyperperiod (VRendVO) = NCS_blocking VRendVO) + NCS_preemption (VRendVO) + NCS_normalExecutionEnd(VRendVO) = 0 + 0 + 8 = 8

43. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 43 NCS Estimation Method Step 14: Total NCS Step 14 Determine the total NCS during the entire execution of the streaming application. We know that the average number of hyperperiods during stable phase HN = 34 (from step 8) => Total estimated NCS: NCS_total(FRead) = 5*34 + 26 + 8 = 204 vs measured 207 NCS_total (VDec) = 9*34 + 22 + 356 = 684 vs measured 679 NCS_total (VRendVO) = 8*34 + 6 + 94 = 372 vs measured 362 Note: Differences come from the fact that we work with averages in the components models which determines an average length for the hyperperiod and an average number of hyperperiods.

44. Philips Research 1st meeting of project EES.5653 29 June 2015 Alina Albu, m.a.albu@tue.nl TU/e Computer Science, System Architecture and Networking Philips Research Laboratories Eindhoven 44 Future work Test method on more complex, realistic case studies Write paper describing the aforementioned findings Extend estimation method for applications containing multiple dependent/independent chains. Continue studies to finding ways to estimate the necessary of memory and bus for streaming applications. Continue studies on estimating necessary of resources streaming applications running on multiple processors platforms.