| Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997. |
....[7] and Di Blasio and Fisher [5] describe two concurrent object calculi, but no account of object migration is given for them. An early version of Emerald [14] includes a form of object migration similar to the one in Obliq, but little formal work is known about it. Finally, in Distributed Oz [34], object migration is a primitive notion, so objects are physically mobile and travel according to a provably safe mobile state protocol from site to site, wherever they are needed or intend to go. 1.4. Outline In Section 2 we introduce the # calculus on which we interpret jeblik. Section 3 ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....been addressed in these works. Sekiguchi and Yonezawa present an encoding of coreObliq into their calculus dist, but they do not consider aliasing [22] Emerald [11] has many things in common with Obliq, including a similar style of migration by means of cloning and aliasing. In Distributed Oz [24] object migration is a primitive notion, so objects are physically mobile and travel according to a provably safe mobile state protocol from site to site, wherever they are needed or intend to go. 2. TYPED PROTECTED SERIALIZED CONCURRENT OBJECTS In the predecessors of this paper, we presented ....
.... ; Aliasing a good idea, after all Since we completely ignored in our study the question of performance, or the feasibility of a run time system to deal with ever growing alias chains, it is hard to compare this style of migration to others that have been proposed, for example for Distributed Oz [24]. These matters go well beyond the scope of this paper, but we may note here that the compression of alias chains should be straightforward based on our formal notion of stable alias nodes. However, we would like to emphasize that the intuitive simplicity of the concept does not quite compensate ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....non determinism does not give rise to unacceptable answers. Obliq [18] supports strong mobility. It has both a graph metaphor and the ability to move demand. Obliq objects are local to the site of creation and therefore marshalling of computation needs to happen at the design stage. The Mozart [129] distributed system implements Oz [115, 45] a concurrent, object oriented language with dataflow synchronisation. Oz combines concurrent and distributed programming [44] with logical constraint based inference, making it a unique choice for developing multi agent systems. Agents represent a ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile Objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....and to get access to native resources. Further Reading An overview on the design of Distributed Oz is [6] A tutorial account on distributed programming with Mozart is [20] The distributed semantics of logic variables is reported in [5] the distributed semantics of objects is discussed in [21]. More information on functors, dynamic linking, and module managers in Mozart can be found in [4] 3 Architecture The concurrent search engine consists of a single manager and several workers. The manager initializes the workers, collects solutions, detects termination, and Manager Worker ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....The guard construct also bears a close relation to that of Orca [4] the key difference being that Orca allows a thread to wait on multiple guards, nondeterministically selecting one if more than one is satisfied. It is also similar to the dataflow synchronization structure of Distributed Oz [40]. The guard keyword is used to mark a block of code as containing an atomic transaction, and to block execution of that transaction until a boolean predicate is satisfied. The boolean predicate must be over the local state of the thread and the fields of the object. Since a guard may be evaluated ....
....actual implementation cannot use such threads, since threads can enter arbitrary (local) Java code. Furthermore, Kan consistency is defined at the object level, for which the DAG approach is inappropriate, since it requires breaking encapsulation to make all dependencies explicit. Distributed Oz [40] is a distributed version of the higher order concurrent constraint language Oz. Oz objects combine stateful data abstraction with mutual exclusion and synchronization, including a dataflow synchronization construct that is similar to Kan s guard statement. It was inspired by concurrent logic ....
Peter van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-- 51, September 1997.
....we focus on the agent mobility only. At the end of Section 3 we discuss possible extensions towards the last, most general model. 1. 3 Distributed concurrent constraint programming To our knowledge, there have been only two previous proposals for distributed extensions of ccp: Distributed Oz [18] and Distributed ccp [13] The proposal in [13] is based on the notion of agents computing within their local stores of constraints, and exchanging constraint abstractions through channels. A process receiving an abstraction applies it to its local variables, thus making a sort of local version ....
....it to its local variables, thus making a sort of local version of the received constraint. The dependency on global information is avoided by a static analysis of the program, giving the sufficient conditions under which the store of two agents can be divided in two local (independent) stores. In [18] the notion of global and local information coexist: the computation of an agent mainly depend on local data, but the bindings on the shared logical variables are global and require handling by a distributed constraint solving algorithm. The main kind of mobility is cell mobility, namely the ....
[Article contains additional citation context not shown here]
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997.
....can be bound to unbound variables, in which case they become identical references. The constraint store is monotonic, i.e. bindings can only be added, not removed or changed. The mutable store consists of mutable references into the constraint store. Mutable references are also called cells [124]. A mutable reference consists of two parts: its name, which is a value, and its content, which is a reference into the constraint store. The mutable store is nonmonotonic because a mutable reference can be changed. The trigger store consists of triggers, which are pairs of variables and ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....and to get access to native resources. Further Reading An overview on the design of Distributed Oz is [48] A tutorial account on distributed programming with Mozart is [150] The distributed semantics of logic variables is reported in [47] the distributed semantics of objects is discussed in [151]. More information on functors, dynamic linking, and module managers in Mozart can be found in [37] 9.3 Architecture The concurrent search engine consists of a single manager and several workers. The manager initializes the workers, collects solutions, detects termination, and assists in ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile Objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile Objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....language interface. We formally define network semantics with its failure models, and we prove that the protocol correctly implements its requirements. Mobile object system. The mobile state protocol implements a cell, an updatable pointer that is the basic stateful entity in the Oz 2 language [8]. The protocol is a fault tolerant extension of the distributed mobile state protocol defined in [8] It has the following design goals. 1) It does not sacrifice performance in the common case of no failures. It has the same performance as the non fault tolerant protocol of [8] 2) When there ....
....the protocol correctly implements its requirements. Mobile object system. The mobile state protocol implements a cell, an updatable pointer that is the basic stateful entity in the Oz 2 language [8] The protocol is a fault tolerant extension of the distributed mobile state protocol defined in [8]. It has the following design goals. 1) It does not sacrifice performance in the common case of no failures. It has the same performance as the non fault tolerant protocol of [8] 2) When there are failures, it provides enough information through its thread interface so that common ....
[Article contains additional citation context not shown here]
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997. 12
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....unbound variables, in which case they become identical references. The constraint store is monotonic, i.e. bindings can only be added, not removed or changed. The mutable store consists of mutable pointers that contain references into the constraint store. Mutable pointers are also called cells [91]. A mutable pointer consists of two parts: its name, which is a value, and its content, which is a reference into the constraint store. The mutable store is nonmonotonic because pointer contents can be changed. The thread store consists of a set of threads. Each thread is de ned by a statement S ....
....In Figure 2, statements are denoted by hSi and hGi, space operations by hSpacei (see Figure 4) and logic variables by italic capitals such as X and Y . The variable P references a procedure and the variable C references a cell. The semantics of the kernel language are given in [74] 68] [91], and [25] Some syntactic sugar is added to Figure 2 to make it more intuitive. In particular, Figure 2 includes three redundant statements, local, if, and case, which can all be expressed with cond. The kernel language splits naturally into ve parts: CORE: The core is strict functional ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997.
....the thread statement and have their own identifier. Exception handling is dynamically scoped and uses the try and raise statements. The full Oz language is defined by transforming all its statements into this kernel language. Oz supports idioms such as objects, classes, reentrant locks, and ports [27, 35]. The system implements them efficiently while respecting their definitions. We will give a brief summary of each idiom s definition. For clarity, we have made small conceptual simplifications. Full definitions are given in [10] ffl Object. An object is essentially a one argument procedure Obj ....
....of four properties: ffl Oz has a simple formal foundation that does not sacrifice expressiveness or efficient implementation. Oz appears to the programmer as a concurrent object oriented language 1 Only ports are changed slightly to better model asynchronous FIFO communication between sites [35]. 4 every bit as advanced as modern languages such as Java. The current emulator based implementation is as good or better than Java emulators [14, 13] Standard techniques for concurrent object oriented design apply to Oz [18] Furthermore, Oz introduces powerful new techniques that are not ....
[Article contains additional citation context not shown here]
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997.
....unbound variables, in which case they become identical references. The constraint store is monotonic, i.e. bindings can only be added, not removed or changed. The mutable store consists of mutable pointers that contain references into the constraint store. Mutable pointers are also called cells [86]. A mutable pointer consists of two parts: its name, which is a value, and its content, which is a reference into the constraint store. The mutable store is nonmonotonic because pointer contents can be changed. The thread store consists of a set of threads. Each thread is defined by a statement S ....
....In Figure 2, statements are denoted by hSi and hGi, space operations by hSpacei (see Figure 4) and logic variables by italic capitals such as X and Y . The variable P references a procedure and the variable C references a cell. The semantics of the kernel language are given in [70] 64] [86], and [25] Some syntactic sugar is added to Figure 2 to make it more intuitive. In particular, Figure 2 includes four redundant statements: local, if, case, and dis. The local, if, and case statements can all be expressed with cond. The dis statement can be programmed with cond and computation ....
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804--851, September 1997. 29
....and Section 6 describes some additional functionality. Section 7 reports our preliminary experimental results, while Section 8 concludes the paper. 2 Overview of CLP( Lin ) In this section, we give an informal overview of CLP( Lin ) The reader interested in more details can refer to [8, 21]. CLP( Lin ) programs have essentially the same syntax as Prolog, the main difference being that CLP( Lin ) programs also allow linear constraints over real numbers to appear in the bodies of clauses. 1 In this article, variables are denoted by uppercase letters, constraints by the letter c ....
P. Van Hentenryck and V. Ramachandran. Backtracking without Trailing in CLP(! lin ). ACM Transactions on Programming Languages and Systems, 1995. (to appear). 16
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997.
No context found.
Peter J. Stuckey and Martin Sulzmann. A theory of overloading. ACM Transaction on Programming Languages and Systems, 2004. To appear.
No context found.
Peter Van Roy, Seif Haridi, Per Brand, Gert Smolka, Michael Mehl, and Ralf Scheidhauer. Mobile objects in Distributed Oz. ACM Transactions on Programming Languages and Systems, 19(5):804-851, September 1997. 13
First 50 documents
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