H. Kopetz and G. Grnsteidl, "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Research Report No. 12/92, Inst. fr Techn. Informatik, Techn. University of Vienna, 1992  Home/Search   Document Not in Database   Summary   Related Articles   Check

....i and sensor S i is housed with a microcontroller of limited computing power and a network interface. Processors exchange messages via a deterministic communication protocol such as time division multiple access (TDMA) having a bounded worst case delivery latency. The timetriggered protocol (TTP) [5] and FlexRay [15] are examples of TDMA based protocols where a dedicated co processor handles communication without interfering with program execution on P i . Processors, sensors, and actuators may suffer a bounded number of operational failures. Permanent faults persist until corrective action ....

H. Kopetz and G. Gruensteidl, "TTP - A Time-Triggered Protocol for Fault-Tolerant Real-Time Systems," Proc. IEEE Fault-Tolerant Comput. Symp., pp. 524-532, 1993.

....for embedded systems that are (i) heavily loaded at run time (and hence require simple scheduling and energymanagement schemes) or (ii) are extremely critical and thus need to be extensively tested to ensure predictability. Such systems may be scheduled with table driven and time triggered [23] schedulers. In addition to providing scheduling information, lookup tables can store the voltage levels that should be supplied to various system components under a static power management scheme, such voltages would be determined o line. Dynamic power management: If a system has the ....

Kopetz, H., and Gr unsteidl, G. TTP - A time-triggered protocol for fault-tolerant real-time systems. In Proceedings of the 23rd Annual International Symposium on Fault-Tolerant Computing (FTCS '93) (Toulouse, France, June 1993), J.-C. Laprie, Ed., IEEE Computer Society Press, pp. 524{ 533.

....redundant transmissions of events and guarantees that the respective publisher has a privileged access to the communication network. Access is based on the reservation of network resources according to a TDMA mechanism (TDMA: Time Division Multiple Access) similar to the time triggered protocol [14]. It means that events published to the HRTEC have to be ready at the start of the assigned TDMA time slot. As explained below, in contrast to most TDMA schemes, we exploit the specific priority mechanisms of the CAN Bus to enforce the temporal guarantees in a more flexible way. A SRTEC has ....

....event channel as an entry in the attribute list. 3 Event channels on a CAN Bus network We now present a mapping of the described abstractions to a field bus network. We will first describe the reservation scheme for the HRTECs which is similar to a scheme used in time triggered protocols like TTP [14], TTP A [15] and TTCAN [7] However, a substantial advantage over a TDMA scheme is that due to CAN Bus properties, bandwidth which was reserved but is not needed by a HRTEC can be used by less critical traffic. Because of the conservative worst case assumptions of a HRTEC, this can be a large ....

[Article contains additional citation context not shown here]

H. Kopetz and G. Grnsteidl. TTP - a time-triggered protocol for fault-tolerant real-time systems. Technical Report 12/92, Inst. fr Techn. Informatik, Techn. University of Vienna, 1992.

....crashed: these protocols are always timely in such synchronous systems because these systems are always majority stable . Another advantage of the timed asynchronous system model is that the protocols designed for timed asynchronous systems are also usable in synchronous systems with blackouts [13]. That is, systems which can be unstable for short intervals but the length and the rate of occurrence of these unstability intervals is bounded. 2 Timed Asynchronous Systems The timed asynchronous system model is described by a set of assumptions on the behavior of processes, communications, and ....

H. Kopetz and G. Grundsteidl. Ttp: A time-triggered protocol for fault-tolerant real-time systems. In Proc. of 23d Symp. on Fault-Tolerant Computing, pages 524--533, Toulouse, France, Jun 1993. 13

....in that we applied the detection schemes to real time communication and evaluated their efficiency. In most previous research on dependable real time communication, even the definition of failures is not clearly given and failures are assumed to be detected immediately upon their occurrence. In [24, 25, 26], the I am alive heartbeat method is implicitly assumed to detect all failures, but this assumption is valid only if the network components have the failure silent characteristic. In general, the failure silent semantic is difficult to achieve without relying on redundancy. Our experimental ....

H. Kopetz and G. Grunsteidl, "TTP - a time-triggered protocol for fault-tolerant real-time systems," in Proc. IEEE FTCS, pp. 524--533, 1993.

....and this information is used to confine propagation. An important issue for co operation is to know what QoS can be sustained by the zones in which participants reside. Typically, a single CAN represents a zone with a very high level of predictability compared to a zone in a wireless network [22,27,36,39,43]. If participants reside in different zones, communication has to adapt to the weakest guarantees including the losses over the gateways. In a mobile environment where migration from one zone to another zone is likely to happen, it is a great challenge to devise communication mechanisms that ....

H. Kopetz and G. Grfinsteidl. TTP - A Time-Triggered Protocol for Fault-Tolerant Real-Time Systems, Research Report 12/92, Inst. f. Techn. Informatik, Tech. Univ. of Vienna, 1992.

....keeping Boeing 777s in the air for nearly a decade but it is also the most expensive: the BIUs provide rich functionality and are fully duplicated at each node. 2.1. 2 TTA The Time Triggered Architecture (TTA) was developed by Hermann Kopetz and colleagues at the Technical University of Vienna [KG93, KG94] Commercial development of the architecture is undertaken by TTTech and it is being deployed for safety critical applications in cars by Audi and Volkswagen, and for flight critical functions in aircraft and aircraft engines by Honeywell. Current implementations of TTA use a bus ....

H. Kopetz and G. Grunsteidl. TTP---a time-triggered protocol for fault-tolerant real-time systems. In Fault Tolerant Computing Symposium 23, pages 524-- 533, Toulouse, France, June 1993. IEEE Computer Society.

....of recognized utility is real time group multicast communication. There are a number of published protocols that deal with the problems of reliable delivery and ordering in synchronous systems, i.e. systems where timeliness bounds (on speed of processors, message transit delay, etc. are known [4, 10]. The problem becomes complex in dynamic and incompletely specified settings, such as the ones targeted by the application classes we are addressing. Indeed, most of previously published work treats such environments under asynchronous assumptions, and thus with no utility for real time. In this ....

H. Kopetz and G. Grunsteidl. TTP - a Time-Triggered Protocol for Fault-Tolerant Real-Time Systems. In Digest of Papers, The 23th International Symposium on Fault-Tolerant Computing, pages 524--533, Toulouse, France, June 1993. IEEE.

....of recognized utility is real time group multicast communication. There are a number of published protocols that deal with the problems of reliable delivery and ordering in synchronous systems, i.e. systems where timeliness bounds (on speed of processors, message transit delay, etc. are known [3, 9]. The problem becomes complex in dynamic and incompletely specified settings, such as the ones targeted by the application classes we are addressing. Indeed, most of previously published work treats such environments under asynchronous assumptions, and thus with no utility for real time. In this ....

H. Kopetz and G. Grunsteidl. TTP - a TimeTriggered Protocol for Fault-Tolerant Real-Time Systems. In Digest of Papers, The 23th International Symposium on Fault-Tolerant Computing, pages 524-- 533, Toulouse, France, June 1993. IEEE.

....control. 1 Introduction Communication protocols take different approaches to achieve order and reliability of message delivery. Reliable multi participant or multicast protocols, for example, lie in two such classes, depending on the use they make of time: ffl Clock driven protocols [4, 6] are diffusion based, and rely on the existence of a clock, in the sense of a global time base that is, an absolute global time reference; ffl Timer driven, or (global ) clock less, protocols [3, 1] are acknowledgement based protocols, that rely on local timers i.e. relative time ....

....paper, g = 0) Since clocks may be as much as apart (the precision) two runs of the protocol from different senders may not have the same delivery time to a given recipient. Steadiness, as discussed in the appendix, is oe = g. Time triggered protocols, such as the TTP protocol described in [6], are clock driven protocols whereby participants only transmit at pre specified periodic instants, driven by the global clock. TTP is thus a TDMA protocol. What the protocol does exactly is to write directly on memory, immediately data arrives at a site, and this, taking into account that every ....

[Article contains additional citation context not shown here]

H. Kopetzand G. Grunsteidl. TTP - a Time-Triggered Protocol for Fault-Tolerant Real-TimeSystems. In Digest of Papers, The 23th International Symposium on Fault-Tolerant Computing, pages 524--533, Toulouse, France, June 1993. IEEE.

....to the IEEE 802.11 standard. So far, there is no protocol providing reliable real time group communication for wireless local area networks based on the IEEE 802.11 standard. There is, however, a number of proposals for achieving reliable real time group communication on wired local area networks [VRR91, RV92, KG93]. None of these protocols is directly suited for our target environment because, firstly, the protocols do not conform to the communication structure imposed by the IEEE 802.11 standard, and secondly, they are not designed to be efficient in the presence of a high degree of message losses. In ....

....KG93] None of these protocols is directly suited for our target environment because, firstly, the protocols do not conform to the communication structure imposed by the IEEE 802.11 standard, and secondly, they are not designed to be efficient in the presence of a high degree of message losses. In [KG93], a static redundancy approach is used to tolerate message losses and meeting real time requirements by sending every message a fixed number of times, independent of a successful transmittal in the meanwhile. This is the so called message diffusion approach that consumes a considerable amount of ....

[Article contains additional citation context not shown here]

H. Kopetz and G. Grnsteidl. TTP -- A Time Triggered Protocol For FaultTolerant Real-Time Systems. In Proceedings of the 23 rd International Symposium on Fault-Tolerant Computing, pp. 524-532, Toulouse, France, 1993.

....by proper off line scheduling of the communication resources. If HRTMs and SRTMs share the same communication medium, it has to be guaranteed that the SRTMs by no means interfere with the hard part. Some static hard scheduling schemes take the approach to consider every message as a periodic HRTM [1, 2]. This method is safe but it does not only consume more resources than actually needed, but also may end up with an infeasible system. Combining static pre planned message scheduling and a dynamic message scheduling would be beneficial. Protocols like the IEEE 802.11 [3] for wireless LAN ....

....message scheduling would be beneficial. Protocols like the IEEE 802.11 [3] for wireless LAN communication, the shared channel concept [4] and recently, the TTP A protocol [5] implement this approach. However, they use a centralised approach with a single master [3] or an ubiquitous global time [2] [4] to guarantee exclusive communication of messages. We did not consider a central master because it constitutes a single point of failure and consumes additional bandwidth. The other approaches require global time in all nodes which has to be provided by a respective protocol. Simple nodes like ....

[Article contains additional citation context not shown here]

H. Kopetz and G. Grnsteidl, "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Research Report No. 12/92, Inst. fr Techn. Informatik, Techn. University of Vienna, 1992

....is indeed correct. However, since such proof efforts require substantial skill and effort, it appears to be very desirable to have available a reusable formal framework that assists in verifying the specific clock synchronization algorithms used in particular practically relevant contexts, such as [8,1,2]. Clocks are synchronized by the periodical application of an adjustment to the local clock value. The required bound between different clock values can be reached either by variation of the re synchronization period length or by variation of the amount of adjustment. Clock synchronization ....

....re integration of clocks. Also, further specific clock synchronization algorithms and convergence functions should be examined as to how well they fit into the general framework presented here. We are currently working on fitting the clock synchronization of the Time Triggered Architecture (TTA) [1,2] into the framework; this requires a slight modification of the theories because of the different way in which the clock values of other nodes are gathered. A further direction of current work is an examination of how the clock synchronization theories can be combined with Rushby s general ....

H. Kopetz and G. Gruensteidl. Ttp -- a time triggered protocol for fault-tolerant real-time systems. IEEE Computer, 27(1):14--23, January 1994.

.... systems such as those for automotive steering, braking, and suspension control [15] It has been argued that the kind of reliability required in such situations cannot be achieved without a careful formal analysis of the mechanisms and algorithms involved [3, 12] The Time Triggered Protocol (TTP) [7] is the core of the communication level of TTA. Group membership is central service of TTP as it provides to all non faulty processors a consistent view of which nodes are operational and which are not at any given moment. Distributed fault tolerant algorithms are inherently difficult to reason ....

H. Kopetz and G. Grnsteidl. TTP -- A Time Triggered Protocol for Fault-Tolerant Real-Time Systems. IEEE Computer, 27(1):14--23, 1994.

....be kept sufficiently close together and that the synchronization be able to tolerate faults to a limited extent. The main purpose of this paper is to present a formal analysis of the clock synchronization algorithm embedded in a specific time triggered context, the TimeTriggered Protocol (TTP) [8, 9]. TTP is the core of the communication level of the Time Triggered Architecture, an architecture that has been developed and evaluated in two recent European projects: Time Triggered Architecture (Esprit OMI program) and X by Wire (Brite EuRam program) The Time Triggered Architecture is ....

H. Kopetz and G. Grunsteidl. TTP -- A Time Triggered Protocol for FaultTolerant Real-Time Systems. IEEE Computer, 27(1):14--23, January 1994.

....in our current prototype Dtp has the same length (500 s) for both soft and hard real time messages, resulting in a constant processing overhead of 2.6 during the transmission of any kind of real time message. The processing overhead of the hybrid bus scheduling approach is similar to that of TDMA [5]. Both protocols need a globally synchronized time reference, and perform the correct bus scheduling using reserved time slot of message h DT h min. gap DGmin critical interval latest start time latest ready time Laxity DL h reserved timeslot of the next message DT fail now Fig. 1 . The ....

H. Kopetz and G. Grünsteidl, "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Research Report No. 12/92, Institut für Technische Informatik, Technical University of Vienna, 1992.

....This means that several subsequent transmissions plus the failure detection recovery of CAN bus, which consumes bounded time (Rufino et al. 1995) have to be scheduled. This application of time redundancy is similar to strategies used in other real time communication protocols, e.g. TTP (Kopetz et al. 1992)) However, in contrast to statically planned communication, this can use the redundant time slot for low priority communication, if the first transmission was successful. 6. CONCLUSION AND FUTURE RESEARCH In order to guarantee timely delivery of hard realtime messages in a CAN network, the ....

Kopetz, H. and G. Grünsteidl (1992). TTP - A TimeTriggered Protocol for Fault-Tolerant Real-Time Systems. Research Report No. 12/92, Inst. für Techn. Informatik, Techn. University of Vienna.

....case of the more general problem to schedule a cooperative distributed computation. There exist several alternative approaches to solve this problem based on the assumptions about the behaviour of the system and the environment. The first approach, known as time triggered approach [ 18] 21] [19]] assumes a complete knowledge of all future actions of the system. Hence, during operation these systems exhibit minimal overhead paired with a maximum of predictability and are highly appropriate for safety critical applications which can be modelled by a periodic behaviour. A more adaptable and ....

.... redundancy to tolerate a single communication failure, we have to schedule two subsequent transmissions plus the fault handling mechanism of the CAN bus, which consumes bounded time [ 29] This strategy is also used in other real time communication protocols, e.g. the Time Triggered Protocol [[19]] However, in contrast to statically planned communication, we can use the redundant time slot for other messages if the first transmission was successful. In the case that a hard real time message is not transmitted before its deadline, it will be discarded, and the sending object is notified. ....

H. Kopetz and G. Grünsteidl: "TTP - A Time-Triggered Protocol for Fault-Tolerant Real-Time Systems", Res. Report 12/92, Inst. f. Techn. Informatik, Technical University of Vienna, 1992.

....is unnecessary, and may lead to bandwidth shortage. In a completely static system, a global calendar is available and each node has its relevant entries referring to its message transmission times in a global time scale. A message may only be transmitted according to this schedule [4] 1] [6]. In a more dynamic system where hard real time, soft real time, and non real time messages coexist, things are more complicated. If the bus is free, a soft real time message may be sent. In this case, it must be guaranteed that it does not cause a timing failure of a hard real time message. 3. ....

H. Kopetz and G. Grünsteidl, "TTP - A Time-Triggered Protocol for Fault-Tolerant Real-Time Systems", Research Report No. 12/92, Inst. für Techn. Informatik, Techn. University of Vienna, 1992.

....ideas of the scheme. Let us start with the assumption for hard real time messages (A2) Hard real time messages are received before their deadlines by all operational nodes under an omission fault assumption. For hard real time messages fixed time slots are reserved like in a TDMA [16] or TTP [17] approach. Conflicting resource requirements between hard real time messages are resolved off line, i.e. at run time two hard real time messages never will compete for the CAN bus. If a time slot for a specific hard real time message arrives, the dynamic priority mechanism assures that the message ....

H. Kopetz and G. Grünsteidl: "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Res. Report 12/92, Inst. f. Techn. Informatik, Tech. Univ. of Vienna, 1992.

....these protocols are always timely in such synchronous systems because these systems are always majority stable 1 . Another advantage of the timed asynchronous system model is that the protocols designed for timed asynchronous systems are also usable in synchronous systems with blackouts [13]. That is, systems which can be unstable for short intervals but the length and the rate of occurrence of these unstability intervals is bounded. 2 Timed Asynchronous Systems The timed asynchronous system model is described by a set of assumptions on the behavior of processes, communications, and ....

H. Kopetz and G. Grundsteidl. Ttp: A time-triggered protocol for fault-tolerant real-time systems. In Proc. of 23d Symp. on Fault-Tolerant Computing, pages 524--533, Toulouse, France, Jun 1993.

....by using redundant components. Due to the very adverse electromagnetic environment, transient faults are by far the most dominant physical phenomenon in automotive systems, especially for data transmission channels. For example, the designers of the TTP protocol for automotive applications [14] assume that transient channel and node failure rates are 10 times greater than the corresponding rates of permanent failure. Furthermore, powerful electromagnetic interference may lead to correlated transient faults or black outs over several sub systems, which may require special global ....

H. Kopetz; G. Grünsteidl : "TTP --- A Time-Triggered Protocol for Fault-Tolerant RealTime Systems", IEEE---23. International Symposium on Fault-Tolerant Computing (FTCS-23), pp. 524-533, June 1993

....among various subsystems (similar to Delta 4) For simplicity, we consider a sender to have failed if a message has not been received correctly by all the receivers. The operation of the system consists of a series of frames with each frame having the same period. During each frame, time triggered [7] as well as event triggered tasks are executed (eventtriggered tasks are handled in the same way as timetriggered tasks, with some potential overhead, by using periodic polling [7] The system can be loosely synchronized at the beginning of each frame using the Shin Dolter algorithm [13] We ....

....of the system consists of a series of frames with each frame having the same period. During each frame, time triggered [7] as well as event triggered tasks are executed (eventtriggered tasks are handled in the same way as timetriggered tasks, with some potential overhead, by using periodic polling [7]) The system can be loosely synchronized at the beginning of each frame using the Shin Dolter algorithm [13] We consider loosely synchronized processors, such as those in a local area network, and loosely synchronous communications. A message should be forwarded to its destination in a bounded ....

H. Kopetz and G. Grunsteidl, "TTP -- A timetriggered protocol for fault-tolerant real-time systems, " 23rd Intl. Symp. on Fault-Tolerant Computing, Toulouse, France, June 1993, pp. 524-533.

....are scarce. The best examples of practical techniques that can be used in constructing hard real time FT DCSs seem to be the PRHB (periodic reception history broadcast) technique formulated by Kopetz, Grunsteidl, and Reisinger in [21] and extended in [19] and its variations, such as the TTP scheme [22]. Hybrid approaches can be developed in rigorous forms only after decentralized approaches are well understood. 17.5 The DRB Scheme for Real Time FT Computing Stations The DRB scheme is one of the most flexible techniques for construction of realtime FT computing stations. The DRB scheme exploits ....

....power needed for RHM analysis and a small part of the bus bandwidth spent for RHV broadcast. The RHM analysis can be off loaded to an auxiliary processor. A similar NSR scheme with less communication overhead, called the time triggered protocol (TTP) was recently formulated by Kopetz in [22]. Further research is needed to find similar efficient NSR schemes that can be implemented on different LAN access schemes, e.g. CSMA with deterministic collision resolution, token ring, and so on. 17.7 Major Remaining Issues (1) Efficient integration of the schemes for construction of real time ....

Kopetz, H. and Grunsteidl, G., "TTP-A: Time Triggered Protocol for FaultTolerant Real-Time Systems," Proc. IEEE Computer Society's FTCS-23, Toulouse, June 1993, pp. 524--533.

....block of a timetriggered architecture. At about that time the important concept of temporal accuracy of real time information was introduced by Kopetz and Kim [23] 24] The TTP C protocol, which includes a clock synchronization service and a membership service, was first published in 1993 [25]. A prototype version of the TTA, including a new clock synchronization chip [26] was built in the context of the European PDCS project. This new prototype implementation has been subject to extensive fault injection experiments [27] 28] From these ex c # 2002 IEEE. To appear in Proceedings ....

....CNI within the interval delimited by these a priori known fetch and delivery instants. The TTA contains two communication protocols that provide this communication service, the fault tolerant TTP C protocol and the low cost fieldbus protocol TTP A. A. The TTP C Protocol The TTP C protocol [25], 52] is a fault tolerant timetriggered protocol that provides the following services: Autonomous fault tolerant message transport with known delay and bounded jitter between the CNIs of the nodes of a cluster by employing a TDMA medium access strategy on replicated communication channels. ....

H. Kopetz and G. Gr unsteidl. TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems. In Proceedings of the 23rd International Symposium on Fault-Tolerant Computing, pages 524--533, 1993.

No context found.

H. Kopetz and G. Grnsteidl, "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Research Report No. 12/92, Inst. fr Techn. Informatik, Techn. University of Vienna, 1992

No context found.

H. Kopetz and G. Grnsteidl. TTP -- A Time Triggered Protocol for Fault-Tolerant Real-Time Systems. IEEE Computer, 27(1):14--23, 1994.

No context found.

H. Kopetz and G. Grnsteidl, "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Research Report No. 12/92, Inst. fr Techn. Informatik, Techn. University of Vienna, 1992

No context found.

H. Kopetz and G. Grunsteidl. TTP - a Time-Triggered Protocol for FaultTolerant Real-Time Systems. In Digest of Papers, The 23th International Symposium on Fault-Tolerant Computing, pages 524--533, Toulouse, France, June 1993. IEEE.

No context found.

H. Kopetz, "TTP - A Time-Triggered Protocol for Fault-Tolerant Real-Time Systems," Proc. IEEE Fault-Tolerant Computing Symp., pp. 524-533, 1993.

No context found.

H. Kopetz and G. Grünsteidl, "TTP - A Time-Triggered Protocol for FaultTolerant Real-Time Systems", Research Report No. 12/92, Institut für Technische Informatik, Technical University of Vienna, 1992.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC