Existing clock synchronization algorithms assume a bounded clock reading error. This, in turn, results in an inflexible design that typically requires node crashes whenever the given bound might be violated. We propose a novel, adaptive internal clock synchronization algorithm which allows to compute the deviation between the clocks during runtime. The computed deviation can be propagated to the application layer to allow it to adapt its behavior according to the current clock deviation. The contributions of this paper are: (1) a new specification of a relaxed clock synchronization problem, and (2) a new clock synchronization algorithm with a novel approach to dealing with crash failures.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
2. However, in timed asynchronous applications:I messages get dropped or are arbitrarily delayedI different nominal and observed oscillator frequenciesI clocks’ speeds fluctuate with timeI processes crash
3. ...thus 1 cannot be guaranteed
Result:
An inherently inflexible and a failure-prone system
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
2. However, in timed asynchronous applications:I messages get dropped or are arbitrarily delayedI different nominal and observed oscillator frequenciesI clocks’ speeds fluctuate with timeI processes crash
3. ...thus 1 cannot be guaranteed
Result:
An inherently inflexible and a failure-prone system
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
2. However, in timed asynchronous applications:I messages get dropped or are arbitrarily delayedI different nominal and observed oscillator frequenciesI clocks’ speeds fluctuate with timeI processes crash
3. ...thus 1 cannot be guaranteed
Result:
An inherently inflexible and a failure-prone system
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
2. However, in timed asynchronous applications:I messages get dropped or are arbitrarily delayedI different nominal and observed oscillator frequenciesI clocks’ speeds fluctuate with timeI processes crash
3. ...thus 1 cannot be guaranteed
Result:
An inherently inflexible and a failure-prone system
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
1 ClockVa l Ap ; // c u r r e n t ad jus tment2 ClockVa l T; // end o f c u r r e n t round
4 vo id i n i t ( ) {5 (Ap ,T) = i n i t i a l A d j u s t m e n t ( ) ;6 // e v e r y P s t a r t i n g at T7 s c h edu l e ( s ynch ron i z e r , P , T) ;8 }
10 vo id s yn ch r on i z e r ( ) {11 // N − number o f p r o c e s s e s12 ClockVa l c l k [N] ,13 ClockVa l e r r [N ] ;14 // remote c l o c k r e a d i n g15 r e adC l o ck s ( c l k , e r r ) ;16 // ad jus tment f o r the next round17 Ap = ad j u s t (Ap , T, c l k , e r r ) ;18 // s e t T to next round19 T = T + P;20 }
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
I Instead we use software clocks:Sp(t) = Hp(t) + ap(t)
I Ap is calculated based on the remote clock readingsI probabilistic remote clock reading [FC99b]I provides values of the remote clocks clk[]I and accompanying remote clock reading errors err[]
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
host A - max sync errhost B - max sync errhost C - max sync errhost D - max sync errhost A - real sync errhost B - real sync errhost C - real sync errhost D - real sync err
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
I Need to build critical systems, however:I underlying components are not synchronousI need to use COTS to cut costI need to use wireless to cut weight
I We propose Adaptive Internal Clock SynchronizationI copes with message delays and process failuresI bounds deviation Ep(T ) from correct clocks in the systemI propagates the deviation Ep(T ) to upper level applications
I Applications use the Ep(T ) to adapt their behaviorI no more synchronization layer crashesI the system is safer
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
I Need to build critical systems, however:I underlying components are not synchronousI need to use COTS to cut costI need to use wireless to cut weight
I We propose Adaptive Internal Clock SynchronizationI copes with message delays and process failuresI bounds deviation Ep(T ) from correct clocks in the systemI propagates the deviation Ep(T ) to upper level applications
I Applications use the Ep(T ) to adapt their behaviorI no more synchronization layer crashesI the system is safer
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
I Need to build critical systems, however:I underlying components are not synchronousI need to use COTS to cut costI need to use wireless to cut weight
I We propose Adaptive Internal Clock SynchronizationI copes with message delays and process failuresI bounds deviation Ep(T ) from correct clocks in the systemI propagates the deviation Ep(T ) to upper level applications
I Applications use the Ep(T ) to adapt their behaviorI no more synchronization layer crashesI the system is safer
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer
Probabilistic internal clock synchronization.In Proceedings of the Thirteenth Symposium on Reliable Distributed Systems (SRDS1994), pages 22–31,October 1994.
Flaviu Cristian and Christof Fetzer.
The timed asynchronous distributed system model.IEEE Trans. on Parallel and Distr. Systems, 10(6):642–657, June 1999.
Christof Fetzer and Flaviu Cristian.
Building fault-tolerant hardware clocks.In Proceedings of the Seventh IFIP Int. Working Conf. on Dependable Computing for Critical Applications,pages 59–78, San Jose, USA, Jan 1999.
Christof Fetzer and Flaviu Cristian.
A fail-aware datagram service.In Iain Bate and Alan Burns, editors, IEE Proceedings - Softw. Eng., volume 146, pages 58–74. IEE, April1999.
Christof Fetzer and Flaviu Cristian.
Fail-awareness: An approach to construct fail-safe systems.Journal of Real-Time Systems, 24(2):203–238, March 2003.
Ute Wappler and Christof Fetzer.
Hardware failure virtualization via software encoded processing.In 5th IEEE Int. Conf. on Industrial Inf. (INDIN 2007), volume 2, pages 977–982, June 2007.
Jennifer Lundelius Welch and Nancy Lynch.
A new fault-tolerant algorithm for clock synchronization.Information and Computing, 77(1):1–36, 1988.
Adaptive Internal Clock Synchronization Zbigniew Jerzak, Robert Fach, and Christof Fetzer