G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc. Middleware 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....cross WAN) call. The following example illustrates the dilemma faced by the user of a typical RPC system. Suppose a linked list is shared among a collection of clients. If searches are frequent, encapsulating the list in an object at a central server is likely to result in poor performance [12, 20]. The alternative is to cache the list at every client and to maintain coherence among the copies, either manually in the application or with the help of object caching middleware [20] But then when one client updates the list, e.g. by inserting an item at the head, the entire list will typically ....

....however, and has only a single coherence model. ScaFDOCS [20] is an object caching framework built on top of CORBA. As in Java RMI, shared objects are derived from a base class and their writeToString and readFromString methods are used to serialize and deserialize internal state. CASCADE [12] is a distributed caching service, structured as a CORBA object. The designers of both [20] and [12] report the problem with CORBA s reference model described in Section 1: every access through the CORBA reference is an expensive cross domain call. LBFS [27] is a low bandwidth network le system ....

[Article contains additional citation context not shown here]

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc., Middleware 2000.

....nested callbacks) while InterWeave allows cache reuse. ScaFDOCS [8] is an object caching framework built on top of CORBA. As in Java RMI, shared objects are derived from a base class and their writeToString and readFromString methods are used to serialize and deserialize internal state. CASCADE [6] is a distributed caching service, structured as a CORBA object. Both ScaFDOCS and CASCADE encounter a fundamental limitation of CORBA s reference model: because everything is an object with no exported data members, every use of a reference parameter incurs a callback. They also suffer from the ....

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc., Middleware 2000.

....these are based on Java. Language independent distributed object systems include PerDiS [16] Legion [18] Globe [42] Microsoft s DCOM, and various CORBA compliant systems. Globe replicates objects for availability and fault tolerance. PerDiS and a few CORBA systems (e.g. Fresco [26] and CASCADE [13]) cache objects for locality of reference. While we speculate that relaxed coherence and views might be applicable to such systems, current implementa tions tend to rely on the inefficient retransmission of entire objects, or the transmission and replay of operation logs. Equally significant from ....

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc., Middleware 2000.

....nested callbacks) while InterWeave allows cache reuse. ScaFDOCS [10] is an object caching framework built on top of CORBA. As in Java RMI, shared objects are derived from a base class and their writeToString and readFromString methods are used to serialize and deserialize internal state. CASCADE [7] is a distributed caching service, structured as a CORBA object. Both [10] and [7] describe a fundamental limitation of CORBA s reference model: because everything is an object with no exported data members, every use of a reference parameter incurs a callback. Krishnaswamy and Haumacher [11] ....

....caching framework built on top of CORBA. As in Java RMI, shared objects are derived from a base class and their writeToString and readFromString methods are used to serialize and deserialize internal state. CASCADE [7] is a distributed caching service, structured as a CORBA object. Both [10] and [7] describe a fundamental limitation of CORBA s reference model: because everything is an object with no exported data members, every use of a reference parameter incurs a callback. Krishnaswamy and Haumacher [11] describe a fast RMI implementation capable of caching objects to avoid serialization ....

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a caching service for distributed CORBA objects. In Middleware, pages 1--23, 2000.

....cross WAN) call. The following example illustrates the dilemma faced by the user of a typical RPC system. Suppose a linked list is shared among a collection of clients. If searches are frequent, encapsulating the list in an object at a central server is likely to result in poor performance [12, 20]. The alternative is to cache the list at every client and to maintain coherence among the copies, either manually in the application or with the help of object caching middleware [20] But then when one client updates the list, e.g. by inserting an item at the head, the entire list will typically ....

....however, and has only a single coherence model. ScaFDOCS [20] is an object caching framework built on top of CORBA. As in Java RMI, shared objects are derived from a base class and their writeToString and readFromString methods are used to serialize and deserialize internal state. CASCADE [12] is a distributed caching service, structured as a CORBA object. The designers of both [20] and [12] report the problem with CORBA s reference model described in Section 1: every access through the CORBA reference is an expensive cross domain call. LBFS [27] is a low bandwidth network file system ....

[Article contains additional citation context not shown here]

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc., Middleware 2000.

....applied to a wide range of applications. It allows the association of different replication consistency protocols with components according to their contexts of (re)use. If many projects have addressed the issue of managing replication in contemp orary middleware (and more specifically in CORBA [3][6] 9] providing configurable replication management in a component based environment is still an open issue. This paper presents our approach to component replication, based on the CORBA Component Model (CCM [8] We show that replication can be managed as a configurable non functional aspect ....

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proceedings of Middleware '00,1-23, April 2000

....a major issue in works on distributed shared memory providing multiple consistency protocols [2] or in mobile databases projects introducing optimistic consistency and application specific reconcilia tion[5] CORBA centered research is also very present with works on flexible caching. The CASCADE [4] project for example is based on the CORBA interceptor mechanism while Flex [10] uses object subclassing and object personalized state capture. The cited projects do certainly consider replication configuration and in some solutions are rather close to the mechanisms used in our experiment. ....

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc. of Middleware'00,1-23, April 2000.

No context found.

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a caching service for distributed CORBA objects. In Proceedings of Middleware '00, pages 1--23, April 2000.

....we start by formally defining a basic set of consistency conditions, in an abstract, implementation independent, way, and show how other consistency conditions can be implemented as a combination of these basic conditions. We then describe how we have implemented these conditions within CASCADE [9], our CORBA object caching service 1 . The paper also describes some interesting optimizations we employed in this implementation. Our specifications and resulting implementation have the following distinct features: Rigor: We provide a formal definition of basic consistency conditions, given ....

....consistency conditions, e.g. sequential consistency. Comprehensiveness and usefulness for applications: The presented specifications cover a wide range of consistency requirements for distributed applications. This is shown by proving that many existing consistency conditions such 1 In [9], we described CASCADE, the motivation behind it, its general implementation and a performance analysis. The current paper is the first place where we formally specify the basic consistency conditions, and elaborate on their exact implementation within CASCADE. as sequential consistency [15] ....

[Article contains additional citation context not shown here]

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proceedings of Middleware '00, pages 1--23, April 2000. The Best Conference Paper award.

No context found.

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc. Middleware 2000.

No context found.

Chockler, G., Dolev, D., Friedman, R., Vitenberg, R.: Implementing a caching service for distributed CORBA objects. In: Middleware'00, Heidelberg, Germany, Springer (2000) 1--23

No context found.

Gregory Chockler, Danny Dolev, Roy Friedman, and Roman Vitenberg. Implementing a caching service for distributed CORBA objects. In Middleware'00, pages 1--23, 2000.

No context found.

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. Proc. of Middleware'00, 2000.

No context found.

G. Chockler, D. Dolev, R. Friedman, and R. Vitenberg. Implementing a Caching Service for Distributed CORBA Objects. In Proc., Middleware 2000.

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