Joachim Niehren. Functional Computation as Concurrent Computation. In POPL '96  Home/Search   Document Details and Download   Summary   Related Articles   Check

....reflected in the calculus. A number of refined type systems for calculi have been studied, addressing polymorphism [FLMR97, LW95, PS97, Tur96, Vas94] directionality [Ode95, PS96] linearity and receptiveness [Ama97, KPT96, San97] deadlock freedom [Kob97] object encodings [San96] confluence [Nie96, NS97], type inference [Gay93, VH93] and other phenomena (this is far from exhaustive) Each allows some particular behavioural discipline of processes to be expressed. It may be useful to contrast typing for calculi with the more standard typing for calculi. A simply typed calculus might have ....

Joachim Niehren. Functional computation as concurrent computation. In Proceedings of the 23rd POPL, pages 333--343. ACM Press, January 1996.

....work has been done on the compilation of whole languages into process calculi, exploring both semantics and expressiveness. Examples include translations between the process calculi CSP and CCS [Li83, Mil87] between the join calculus and the # calculus [FG96] of # calculi into process calculi [Mil92, San94b, Lav93, San94a, Tho95, San94b, San95a, ALT95, Ode95b, Nie96], data types and other sequential programming constructs [Mil89, Mil93, Wal91b, Ode95a] from object oriented languages into process calculi [Vaa90, Wal91a, Wal95, Wal93, Jon93, Wal94, PT95] from logic programming languages into the # calculus [Ros93, Li94] and from concurrent constraint ....

J. Niehren. Functional Computation as Concurrent Computation. In J. G. Steele, ed, Proceedings of POPL '96, pages 333--343. ACM, Jan. 1996. Full version as Research Report RR-95-14, DFKI Saarbrucken.

....interaction between sequential execution (including backtracking and cut pruning) and coroutining are made precise. In particular cases where this interaction can lead to indeterministic results are exhibited. Functional computation can be seen as a special form of concurrent computation. In [67], a uniformly confluent core of the calculus is presented which is also contained in models of higher order concurrent constraint programming. The call by need and the call by value calculus can be embedded into the calculus. Furthermore, the call by need complexity is dominated by call byvalue ....

....constraint programming. The call by need and the call by value calculus can be embedded into the calculus. Furthermore, the call by need complexity is dominated by call byvalue complexity. In contrast to the recently proposed call by need calculus, the concurrent call by need model of [67] incorporates mutual recursion and can be extended to cyclic data structures by means of constraints. 68] investigates concurrency as unifying computational paradigm which integrates functional, constraint, and object oriented programming. The ae calculus is proposed as a foundation of ....

Joachim Niehren. Functional computation as concurrent computation. In Symposium on Principles of Programming Languages, St Petersburg Beach, FL, 21--24 January 1996. ACM. To appear.

....as logical entailment between constraints. We extend ccp with higher order procedures (i.e. first class procedures) The resulting higherorder ccp model is simple and theoretically clean. Its restriction to the pure functional paradigm recovers nice formal properties such as confluence [Nie96], both in the lazy and eager cases. For a deeper introduction, see the Oz programming model [Smo95a] and the calculi underlying Oz [Smo94, NM95b, Nie96] A closely related model of concurrency is the calculus [Mil93] See [Nie96] for a formal comparison between the two models. 2 The ....

....higherorder ccp model is simple and theoretically clean. Its restriction to the pure functional paradigm recovers nice formal properties such as confluence [Nie96] both in the lazy and eager cases. For a deeper introduction, see the Oz programming model [Smo95a] and the calculi underlying Oz [Smo94, NM95b, Nie96]. A closely related model of concurrency is the calculus [Mil93] See [Nie96] for a formal comparison between the two models. 2 The non declarative aspect has received some attention, e.g. Nai86, HL94, PS95, And95] 3 The higher order ccp core is minimally extended with state and search ....

[Article contains additional citation context not shown here]

Joachim Niehren. Functional Computation as Concurrent Computation. In 23 rd ACM Symposium on Principles of Programming Languages, 1996.

....extension of the calculi to deal with recursive lets leads to a loss of con uence [Je rey 1993; Ariola and Klop 1997] The original call by need calculus considers recursive lets only brie y. To recover con uence, one can simply disallow reductions under cycles, as in e.g. Benaissa et al. 1996; Niehren 1996]. Ariola and Blom give a full study of cyclic recursion in [Ariola and Blom 1997; Ariola and Blom 1998] and show that an approximation to con uence can be obtained by equating terms with the same in nite normal form. Their share calculus can be seen as the natural successor to the call by need ....

Niehren, J. 1996. Functional computation as concurrent computation. In Proc. POPL'96, the 23 rd ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages (Jan. 1996), pp. 333-343. ACM Press.

....of Section 6 is new. The uniform encoding in Section 7 is from [OD93] A study of the correctness of the call by need encoding in Table 7 is in [BO95] Encodings of graph reductions, related to call by need, into calculus are given in [Bou94, Jef93] but their correctness is not studied. Niehren [Nie96] uses encodings of call byname, call by value and call by need calculi into calculus to compare the time complexity of the strategies. The encoding of Table 8 is precisely Milner s original encoding of N . The optimisation (30) is from [MS98] An encoding of the rule into calculus was rst ....

J. Niehren. Functional computation as concurrent computation. In Proc. 23th POPL. ACM Press, 1996.

....work has been done on the compilation of whole languages into process calculi, exploring both semantics and expressiveness. Examples include translations between the process calculi CSP and CCS [Li83, Mil87] between the join calculus and the calcu lus [FG96] of calculi into process calculi [Mil92, San94b, Lav93, San94a, Tho95, San95a, ALT95, Ode95b, Nie96], data types and other sequential programming constructs [Mil89, Mil93, Wal91b, Ode95a] from object oriented languages into process calculi [Vaa90, Wal91a, Wal95, Wal93, Jon93, Wal94, PT95] from logic programming languages into the calculus [Ros92, Li94] and from concurrent constraint ....

J. Niehren. Functional Computation as Concurrent Computation. In Proceedings of POPL '96, pages 333--343. ACM, Jan. 1996.

....Nestmann [SN95] for the purpose of analyzing confluence in pi calculus processes arising in the semantics of concurrent object oriented programs. The same problem has been tackled by Liu and Walker using purely semantic techniques [LW95a] Our analysis is also related to the one used by Niehren [Nie96] to guarantee uniform confluence in a pi calculus fragment with only replicated inputs. Effects similar to our linear typing have also been achieved by more syntactic means, as in Honda and Yoshida s notion of beta reduction for concurrent combinators [HY94] A more recent paper by Takeuchi, ....

Joachim Niehren. Functional computation as concurrent computation. In Principles of Programming Languages, January 1996.

.... lost once lambda abstraction and cycles are admitted, unless the theory is powerful enough to represent irregular structures as shown in [6, 7] To regain confluence, current formulations of cycles either impose restrictions, such as disallowing reduction under a lambda abstraction or on a cycle [6, 7, 15, 31, 35], or adopt a framework based on interaction nets [24] As discussed in [28] and [12] cycles do not destroy confluence in the context of interaction nets, but only at the expense of greater complexity. In this paper, we limit our attention to cyclic lambda graphs that occur in current ....

J. Niehren. Functional computation as concurrent computation. In Proc. ACM Conference on Principles of Programming Languages, pages 333--343, 1996.

....as logical entailment between constraints. We extend ccp with higher order procedures (i.e. first class procedures) The resulting higher order ccp model is simple and theoretically clean. Its restriction to the pure functional paradigm recovers nice formal properties such as confluence [28], both in the lazy and eager cases. For a deeper introduction, see the Oz programming model [41] and the calculi underlying Oz [40, 29, 28] A closely related model of concurrency is the calculus [25] See [28] for a formal comparison between the two models. The higher order ccp core is ....

....resulting higher order ccp model is simple and theoretically clean. Its restriction to the pure functional paradigm recovers nice formal properties such as confluence [28] both in the lazy and eager cases. For a deeper introduction, see the Oz programming model [41] and the calculi underlying Oz [40, 29, 28]. A closely related model of concurrency is the calculus [25] See [28] for a formal comparison between the two models. The higher order ccp core is minimally extended with state and search (see Section 5) The programming style arising from the resulting model can be sketched as follows: ....

[Article contains additional citation context not shown here]

Joachim Niehren. Functional Computation as Concurrent Computation. In 23 rd ACM Symposium on Principles of Programming Languages, 1996.

.... and multi paradigm languages based on the concurrent constraint paradigm, such as AKL [21] and Oz [35] achieve co ordination in much the same way as Goffin, but functional computations are considered to be a restricted form of relational computations, and so, functions are realized by predicates [28]. As a result there is no distinction between co ordination and computation. Furthermore, AKL and Oz lack a strong type system, an important feature of modern functional programming languages [3] In this context, it is important to note that it is the type system, which makes the static ....

Joachim Niehren. Functional computation as concurrent computation. Research Report RR-95-14, Deutsches Forschungszentrum fur Kunstliche Intelligenz, Stuhlsatzenhausweg 3, D66123 Saarbrucken, Germany, 1995. A shorter version appeared in: Proceedings of POPL'96 , The ACM Press.

....calculi embeddings. OzFun OzFun calculus Compiler Oz ae calculus The relationship between programming languages and the corresponding calculi is given by a simple syntax transformation. The calculi yield a programming model (the OzFun calculus [10] and an implementation model (the ae calculus [15, 16]) OzFun is designed in a way that it has got a very simple programming model. The OzFun calculus models termination, result und runtime errors of mixed eager and lazy functional computation. It is merely a combination of the eager and lazy calculus [21] The embedding of the OzFun calculus into ....

Joachim Niehren. Functional computation as concurrent computation. In Proceedings of the ACM Symposium on Principles of Programming Languages. ACM Press, January 1996. to appear.

....Nestmann [SN95] for the purpose of analyzing confluence in pi calculus processes arising in the semantics of concurrent object oriented programs. The same problem has been tackled by Liu and Walker using purely semantic techniques [LW95a] Our analysis is also related to the one used by Niehren [Nie96] to guarantee uniform confluence in a pi calculus fragment with only replicated inputs. 8 Variants and Extensions We have focused here on a fragment of the original polyadic calculus of [Mil91] omitting output guards (processes of the form x [ y] P ) and the choice operator . Although ....

Joachim Niehren. Functional computation as concurrent computation. In Principles of Programming Languages, January 1996.

....more work has been done on the compilation of whole languages into process calculi, exploring both semantics and expressivity. Examples include translations be tween the process calculi CSP and CCS [Li83, Mil87] between the join calculus and the calculus [FG96] of calculi into process calculi [Mil90, San92, Lav93, San94a, Tho95, San95a, ALT95, Ode95b, Nie96], data types and other sequential programming constructs [Mil89, Mil91, Wal91b, Ode95a] from object oriented languages into process calculi [Vaa90, Wal91a, Wal92, Wal93, Jon93, Wal94, PT95] from logic programming lan guages into the calculus [Ros92, Li94] and from concurrent constraint ....

Joachim Niehren. Functional Computation as Concurrent Computation. In POPL '96

....reflected in the calculus. A number of refined type systems for calculi have been studied, addressing polymorphism [FLMR97, LW95, PS97, Tur96, Vas94] directionality [Ode95, PS96] linearity and receptiveness [Ama97, KPT96, San97] deadlock freedom [Kob97] object encodings [San96] confluence [Nie96, NS97], type inference [Gay93, VH93] and other phenomena (this is far from exhaustive) Each allows some particular behavioural discipline of processes to be expressed. It may be useful to contrast typing for calculi with the more standard typing for calculi. A simply typed calculus might have ....

Joachim Niehren. Functional computation as concurrent computation. In Proceedings of the 23rd POPL, pages 333--343. ACM Press, January 1996.

No context found.

Joachim Niehren. Functional computation as concurrent computation. In Proceedings of the ACM Symposium on Principles of Programming Languages. The ACM Press, January 1996.

....programming and the p calculus includes the following. Smolka [36] presents OPM in form of the g calculus, and notices a close relation to the p calculus. Niehren and Muller [27] prove this relation based on the r calculus which extends the g calculus with a parametric constraint system. Niehren [25, 26] studies uniform confluence in the concurrent calculi mentioned above (g, d, r, p, join) he identifies a common subcalculus of g and p and proves that it can embed the eager and the call by need l calculus. Victor and Parrow [40] give an embedding of the g calculus into the p calculus and prove ....

J. Niehren. Functional computation as concurrent computation. In 23 Symp. on Principles of Programming Languages, pp. 333--343, 1996. ACM Press.

....leads to a transparently distributed constraint store. From this point of view, combination can be interpreted as unification which may or may not involve a network transfer. Related Work: Most surprisingly, the lazy calculus can be embedded into ae( with call by need complexity (see [17]) Alternatively, the call by need calculus [2] could be directly embedded into ae( Both results are stronger than the analogous results for [4] since both embeddings map into a uniformly We owe the idea to personal communication with Gert Smolka and Martin Odersky. Indeterminism via ....

....cells and conditionals. If (E i ) i=0 is maximal, then (F i ) i=0 must be maximal. Otherwise, we could contradict maximality of (E i ) i=0 by applying our bisimulation the other way around. 2 5 Uniformly Concurrent Subcalculi Functional programming is a special form of concurrent programming [17]. Result, termination, and complexity of functional programs are independent of the execution order. For eager functional programming this is reflected by the eager calculus, and for lazy functional programming by the call by need calculus [2] Concurrent computation satisfying the above three ....

[Article contains additional citation context not shown here]

J. Niehren. Functional computation as concurrent computation, 1995. Submitted, http://ps-www.dfki.uni-sb.de/niehren.

....complexity ) This implementation can be organized incrementally. Type Analysis. One application for INES constraints which we are investigating in [23] is type analysis for concurrent constraint programming [17, 27] in particular Oz [28] As formal foundations we intend to use the calculi in [24, 25]. There, INES constraints are used to approximate the set of run time values for program variables. Since values in Oz include infinite trees, it is important that INES allows an interpretation over sets of possibly infinite trees. It is considered an error if the set of possible run time values ....

J. Niehren. Functional Computation as Concurrent Computation. In 23 POPL, pp. 333--343. ACM, 1996.

....justify important optimizations has a long tradition in constraint (logic) programming. The most prominent example is the change from finite to rational trees in Prolog II [14] others are constraints over strings [15] and over Booleans [11] programming languages [29, 38] such as Oz [40] see [32, 33] for formal foundations) In these languages, execution proceeds by adding conjuncts to a global constraint store which grows monotonically during the execution of a program. The addition of an inconsistent constraint is considered a programming error. 4 Consider the following Oz program. The ....

J. Niehren. Functional computation as concurrent computation. In 23 th Proceedings of the ACM Symposium on Principles of Programming Languages, pages 333--

No context found.

Joachim Niehren. Functional Computation as Concurrent Computation. In POPL '96

No context found.

Joachim Niehren. Functional computation as concurrent computation. In Principles of Programming Languages, January 1996.

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