C. Fetzer and F. Cristian. Fortress: A system to support fail-aware realtime applications. In IEEE Workshop on Middleware for Distributed RealTime Systems and Services, San Francisco, Dec 1997. http://www.cs.ucsd.- edu/cfetzer/FORTRESS.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....describes the fail aware datagram service that is central to failure suspicion, and [Fetzer99] the leader election protocol. The applicability of timed model to building real time applications are argued in [Fetzer97a] using traffic control as an example application. The Fortress system [Fetzer97b] implements the basic services of this model and enables real time applications to be built. 2.6. Concluding Remarks We have presented the synchronous and the asynchronous models of distributed systems. In the restricted synchronous model, fault tolerant applications can be built for all fault ....

C. Fetzer and F. Cristian, "Fortress: A System to support Fail Aware Real-Time Applications", In IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997. http://www.cs.ucsd.edu/~cfetzer/Fortress.

....all our measurements on a cluster of 9 Sun IPX workstations connected by a 10Mbit Ethernet in our Dependable Systems Laboratory at UCSD. Seven of these computers run SunOS 4.1.2 while 2 machines run Solaris 2.5. The measurement programs use different services provided by the FORTRESS toolkit [18]. FORTRESS uses UDP for interprocess communication. 1.8 2.3 2.8 3.3 3.8 4.3 250 750 1250 1750 bytes ms Fig. 8. Measured minimum delay of round trip message pairs for different message sizes. We used 20,000 round trips for each of the 156 measured message sizes. To model the dependence ....

C. Fetzer and F. Cristian. Fortress: A system to support fail-aware realtime applications. In IEEE Workshop on Middleware for Distributed RealTime Systems and Services, San Francisco, Dec 1997. http://www.cs.ucsd.- edu/cfetzer/FORTRESS.

....relation is actually symmetric for any two processes (like ) that are in two different 1 partitions. 4 Fail Aware Services We have designed and implemented a hierarchy of fail aware services to support the design and implementation of fail safe real time applications (see Figure 6) [11]. The foundations of the hierarchy are an asynchronous datagram service and a process management service that provide the semantics assumed by the timed asynchronous system model, i.e. messages have omission performance and processes have crash performance failure semantics. We give an overview ....

C. Fetzer and F. Cristian. Fortress: A system to support fail-aware real-time applications. In IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997. http://www.christof.org/- FORTRESS.

....Figure 16: Fortress is a system that supports the development of fault tolerant real time applications. All services it provides are fail aware, i.e. each server lets its clients know whenever it cannot provide its standard semantics. 8 Fortress In this section we describe the Fortress system [11] which is an attempt to provide a common framework for designing real time applications for systems in which the failure rate cannot be bounded a priori. The main issues we address in Fortress are (1) the support of redundancy management to allow the masking of crash, performance, and omission ....

....to adapt the quality of service. Failawareness is a concept that detects all non maskable failures of a server and propagates that information to the clients of the server using indicators. We introduced the notion of fail awareness in [7] A more detailed description of Fortress can be found in [11] and an overview of fail awareness in a partitionable setting is given in [10] The fail aware services is are given in [8, 12, 9] A simple synchronized traffic signaling example that demonstrates the use of fail aware services in a partitionable environment is described in [10] and a more ....

[Article contains additional citation context not shown here]

C. Fetzer and F. Cristian. Fortress: A system to support fail-aware real-time applications. In IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997. 13

....in each network partition there exists an access to a reference time provider, such as a GPS receiver [6] one cannot guarantee that the deviation between all clocks is bounded. We actually use the fail aware datagram service in a fail aware clock synchronization algorithm to read remote clocks [13]. The fail aware datagram service is the foundation of all other fail aware services we have designed and implemented (see Figure 4; 13] It has been vital in the design of all theses services because they all need to reject slow messages. We use the fail aware datagram service in partitionable ....

....deviation between all clocks is bounded. We actually use the fail aware datagram service in a fail aware clock synchronization algorithm to read remote clocks [13] The fail aware datagram service is the foundation of all other fail aware services we have designed and implemented (see Figure 4; [13]) It has been vital in the design of all theses services because they all need to reject slow messages. We use the fail aware datagram service in partitionable systems to provide the abstraction of clean partitions in the sense that, even when slow messages from other partitions arrive, higher ....

[Article contains additional citation context not shown here]

C. Fetzer and F. Cristian. Fortress: A system to support fail-aware real-time applications. In IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997. http://www.christof.org/FORTRESS.

....in each network partition there exists an access to a reference time provider, such as a GPS receiver [6] one cannot guarantee that the deviation between all clocks is bounded. We actually use the fail aware datagram service in a fail aware clock synchronization algorithm to read remote clocks [13]. The fail aware datagram service is the foundation of all other fail aware services we have designed and implemented (see Figure 4; 13] It has been vital in the design of all theses services because they all need to reject slow messages. We use the fail aware datagram service in partitionable ....

....deviation between all clocks is bounded. We actually use the fail aware datagram service in a fail aware clock synchronization algorithm to read remote clocks [13] The fail aware datagram service is the foundation of all other fail aware services we have designed and implemented (see Figure 4; [13]) It has been vital in the design of all theses services because they all need to reject slow messages. We use the fail aware datagram service in partitionable systems to provide the abstraction of clean partitions in the sense that, even when slow messages from other partitions arrive, higher ....

[Article contains additional citation context not shown here]

C. Fetzer and F. Cristian. Fortress: A system to support fail-aware real-time applications. In IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997. http://www.christof.org/FORTRESS.

....all our measurements on a cluster of 9 Sun IPX workstations connected by a 10Mbit Ethernet in our Dependable Systems Laboratory at UCSD. Seven of these computers run SunOS 4.1.2 while 2 machines run Solaris 2.5. The measurement programs use di erent services provided by the FORTRESS toolkit [27]. FORTRESS uses UDP for interprocess communication. To model the dependence of message transmission times on the message size (see Figure 8) we could replace constants and min by two functions that increase with the size of a message. We actually use in [19] such a function for min , i.e. ....

C. Fetzer and F. Cristian, \Fortress: A system to support failaware real-time applications," in IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997, http://www.christof.org/FORTRESS.

....all our measurements on a cluster of 9 Sun IPX workstations connected by a 10Mbit Ethernet in our Dependable Systems Laboratory at UCSD. Seven of these computers run SunOS 4.1.2 while 2 machines run Solaris 2.5. The measurement programs use different services provided by the FORTRESS toolkit [18]. FORTRESS uses UDP for interprocess communication. To model the dependence of message transmission times on the message size (see Figure 8) we could replace constants ffi and ffi min by two functions that increase with the size of a message. We actually use in [19] such a function for ffi min ....

C. Fetzer and F. Cristian. Fortress: A system to support failaware real-time applications. In IEEE Workshop on Middleware for Distributed Real-Time Systems and Services, San Francisco, Dec 1997. http://www.cs.ucsd.edu/~cfetzer/FORTRESS.

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