P. Verssimo and C. Almeida, "Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models, " Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), vol. 7, no. 4, pp. 35--39, Winter 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....that contains as an upper bound for the tranmission delay of . We introduced in [19] several optimizations to provide a better upper bound for . 12] describes the use of the timed model and a fail aware datagram service in a fully automated train control system. The quasi synchronous model [31] is another approach to define a model that is in between synchronous systems and timefree asynchronous systems. It requires (P1) bounded and known processing speeds, P2) bounded and known message delivery times, P3) bounded and known drift rates for correct clocks, P4) bounded and known load ....

P. Verissimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. IEEE TCOS Bulletin, 7(4), Dec 1995.

....upper bound) is negligible. We use the timed asynchronous distributed system model as the foundation of our work because it does not put any bound on the number of failures per time unit. For a detailed, formal description and comparison with other models like the quasi synchronous model of [24] see [5] 2.2 Timed Asynchronous Systems The timed asynchronous system model [5] assumes that processes have access to a local unsynchronized hardware clock with a bounded drift rate, i.e. the clock proceeds within a linear envelope of real time. It uses the following failure model: processes ....

....according to that paradigm are Mars [18] XPA [25] TTP [17] and the Advanced Automation System family [4] Recently, some of the research has focused on adaptive real time systems. Research on detecting performance failures in quasi synchronous systems is described by Almeida and Verissimo [24]. Their approach depends on the existence of a lower level synchronous communication channel to detect such failures. In contrast, our approach does not require such a basic channel, but uses unsynchronized local clocks with bounded drift rates to detect performance failures. There also exist at ....

P. Verissimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. IEEE TCOS Bulletin, 7(4), Dec 1995.

....is impossible for a deterministic algorithm if even one party crashes [FLP85] Two ways to solve it nevertheless have emerged: Weakening the Model. Various models have been proposed for a system that behaves realistically, but that o ers enough synchrony to solve the Byzantine agreement problem [DLS88, VA95, CF95]. For most recent implementations, the failure detector approach [CT91] has been chosen. Most practical protocols implemented in a failure detector model deal only with crash failures, as failure detectors in this model are much easier to handle. Recently, several groups started moving the ....

P.Verissimo and C. Almeida. Quasi-synchronism: a step away from the traditional faulttolerant real-time system models. Winter 1995 Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS ), 7 (4):35 - 39, 1995. 9

....partial synchronous models. Research in distributed systems has traditionally been divided between two canonical models: fully synchronous and fully asynchronous. Partial synchronous models try to give the best of both worlds, allowing timeliness specifications but accepting that they can fail [22, 6]. The TTCB provides a set of time services whose main objective is precisely to detect the failure of timeliness specifications. This is only possible because the TTCB is timely, i.e. the TTCB is a real time (synchronous) component. Many of the time related ideas and services of the TTCB were ....

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bullettin of the Technical Committee on Operating Systems and Application Environments, 7(4):35--39, 1995.

....messages appropriately, we divide the system into parts and assure that at least one of them will be timely. In other words, our assumed model of computation can be considered synchronous but only with respect to the process that transmits the highest priority message. Other models of computation [14, 15] have assumed different levels of synchrony in different parts of the system. These models consider the system made of two distinct parts, one synchronous and the other semi synchronous. The synchronous part is used to control and adjust the semisynchronous one, where complete synchronism may not ....

P. Verssimo and C. Almeida. "Quasi-Synchronism: A Step Away from the Traditional Fault-Tolerant RealTime Systems Model". In Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS 95), pages 35--99, 1995.

....that contains m as an upper bound for the tranmission delay of m. We introduced in [19] several optimizations to provide a better upper bound for m. 20] describes the use of the timed model and a fail aware datagram service in a fully automated train control system. The quasi synchronous model [21] is another approach to de ne a model that is in between synchronous systems and time free asynchronous systems. It requires (P1) bounded and known processing speeds, P2) bounded and known message delivery times, P3) bounded and known drift rates for correct clocks, P4) bounded and known load ....

P. Verissimo and C. Almeida, \Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models," IEEE TCOS Bulletin, vol. 7, no. 4, Dec 1995.

.... oracle that ensures time domain correctness of applications in environments of uncertain synchronism [Verissimo et al. 2000] The timely computing base model addresses a broader spectrum of problems than those solved by previous timed partially synchronous models, such as the quasi synchronous [Verssimo Almeida 1995] and the timed asynchronous models [Cristian Fetzer 1998] All these works share the same observation: synchronism or asynchronism are not homogeneous properties of systems. That is, they vary with time, and they vary with the part of the system being considered. However, each model has treated ....

P. Verssimo and C. Almeida, "Quasi-Synchronism: a Step Away from the Traditional Fault-Tolerant Real-Time System Models", Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7 (4), pp.35-9, 1995. Reference Model and Use Cases 112

....represent various data domain faults as a state transition by defining the rates at which the fault switches states, a time variable capturing the duration of the fault [18] Figure 2 depicts the state transition diagram. 1 Other system models such as timed asynchronous [8] and quasi synchronous [39] also appear in the literature. 5 Benign No Fault Fault Active Fault c(t) d(t) b(t) a(t) Figure 2: Data Domain Fault Models The rates at which permanent, transient and intermittent faults switch states are as follows: ffl Permanent: a(t) 0; b(t) c(t) d(t) 0 ffl Transient: a(t) ....

P. Verissimo, C. Almeida, "Quasi-synchronism: A Step Away from the Traditional Fault-Tolerant Real-Time System Models." IEEE Bulletin of the TCOS, 7(4), pp. 35--39, Winter 1995.

....that contains m as an upper bound for the tranmission delay of m. We introduced in [19] several optimizations to provide a better upper bound for m. 12] describes the use of the timed model and a fail aware datagram service in a fully automated train control system. The quasi synchronous model [31] is another approach to define a model that is in between synchronous systems and time free asynchronous systems. It requires (P1) bounded and known processing speeds, P2) bounded and known message delivery times, P3) bounded and known drift rates for correct clocks, P4) bounded and known load ....

P. Verissimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. IEEE TCOS Bulletin, 7(4), Dec 1995.

....support for real time applications. For example, TTP [9] RTCAST [10] and XPA [11] provide guaranteed response time by exploiting real time operating system features or special purpose hardware support. Other approaches including the fail awareness framework [12] quasi synchronous model [13], and Cactus [14] are based on the best effort paradigm. One of the primary motivations for group communication is its use as a building block for providing replication and consistency in distributed applications. As the needs of distributed applications have grown, new protocols and services have ....

P. Verissimo and C. Almeida, "Quasi-synchronism: A step away from the traditional fault-tolerant real-time system model," IEEE TCOS Bulletin, vol. 7, no. 4, , December 1995.

No context found.

P. Verssimo and C. Almeida, "Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models, " Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), vol. 7, no. 4, pp. 35--39, Winter 1995.

No context found.

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the TCOS, 7(4):35--39, Winter 1995.

.... behavior, and where hardware clocks provide sufficient synchronism to make decisions such as detection of timing failures or fail safe shutdown [8] the quasi synchronous model, where parts of the system have enough synchronism to perform realtime actions with a certain probability[20]. All these works share a same observation: synchronism or asynchronism are not homogeneous properties of systems. That is, they vary with time, and they vary with the part of the system being considered. However, each model has treated these asymmetries in its own way: some relied on the ....

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7(4):35--39, Winter 1995.

....from soft to hard real time. These are challenging problems in largescale systems with uncertain synchrony, especially where wireless communication is employed. We intend to use and build on previous results on partial synchrony systems, such as the timed asynchronous and quasi synchronous models [11,37,40]. 2.4. Scalability Scalability represents a crucial transparency property concerning the ability to accommodate growth in a large scale distributed system. Thus, connecting more participants to the system dynamically, including adding entire additional networks, or providing new services, should ....

P. Verissimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models, Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7(4):35-39, Winter 1995.

.... behavior, and where hardware clocks provide sufficient synchronism to make decisions such as detection of timing failures or fail safe shutdown [8] the quasi synchronous model, where parts of the system have enough synchronism to perform realtime actions with a certain probability[20]. All these works share a same observation: synchronism or asynchronism are not homogeneous properties of systems. That is, they vary with time, and they vary with the part of the system being considered. However, each model has treated these asymmetries in its own way: some relied on the ....

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7(4):35--39, Winter 1995.

.... model, where hardware clocks provide sufficient synchronism to make decisions such as detection of timing failures or fail safe shutdown [6] the quasi synchronous model, where some parts of the system have enough synchronism to perform real time actions with a certain probability [17], and the work on partial synchrony presented in [7, 8] All these papers have in fact motivated the idea behind the work on the TCB model [19] the search of a generic paradigm for systems with uncertain temporal behavior. The significance of the TCB model in the context of factory environments ....

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7(4):35-- 39, Winter 1995.

....fully synchronous, which is the case of these new infrastructures. Some new communication network technologies (e.g. ATM [11] have improved the synchronism properties of these distributed environments, however they are not always fully synchronous. They are at most what we call quasi synchronous [26]. Only a small part of the system can be considered as synchronous. The rest has a more dynamic behavior exhibiting, for a given activity, worst case delays that are much higher than the normal delays 1 (see Figure 1) probability timing failures time D D max T min T Figure 1. ....

....real time applications is a difficult task. However, there is a demand for such applications, and therefore providing support for their development is of utmost importance. We have been addressing these problems and in previous documents we have presented: the quasi synchronous model [26]; group communication protocols that take into account the characteristics of these systems and provide message early delivery [2] and a timing failure detection service that has a key role in the proposed architecture [3] In this paper we describe the protocols associated with the management of ....

[Article contains additional citation context not shown here]

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), IEEE Computer Society, 7(4):35--39, Winter 1995.

.... timing failure detection oracle that ensures time domain correctness of applications in environments of uncertain synchronism[14] The timely computing base model addresses a broader spectrum of problems than those solved by previous timed partially synchronous models, such as the quasi synchronous[12] and the timedasynchronous models[4] All these works share the same observation: synchronism or asynchronism are not homogeneous properties of systems. That is, they 5 vary with time, and they vary with the part of the system being considered. In the timely computing base model, it is assumed ....

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7(4):35-39, Winter 1995.

.... timing failure detection oracle that ensures time domain correctness of applications in environments of uncertain synchronism[14] The timely computing base model addresses a broader spectrum of problems than those solved by previous timed partially synchronous models, such as the quasi synchronous[12] and the timedasynchronous models[4] All these works share the same observation: synchronism or asynchronism are not homogeneous properties of systems.Thatis,they 5 vary with time, and they vary with the part of the system being considered. In the timely computing base model, it is assumed that ....

P. Verssimo and C. Almeida. Quasi-synchronism: a step away from the traditional fault-tolerant real-time system models. Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), 7(4):35--39, Winter 1995.

No context found.

P. Verssimo and C. Almeida. "Quasi-Synchronism: A Step Away from the Traditional Fault-Tolerant Real-Time Systems Model". In Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS), pages 35--99, 1995.

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