Modern generational garbage collectors look for garbage among the young objects, because they have high mortality; however, these objects include the very youngest objects, which clearly are still live. We introduce new garbage collection algorithms, called age-based, some of which postpone consideration of the youngest objects. Collecting less than the whole heap requires write barrier mechanisms to track pointers into the collected region. We describe here a new, efficient write barrier implementation that works for age-based and traditional generational collectors. To compare several collectors, their configurations, and program behavior, we use an accurate simulator that models all heap objects and the pointers among them, but does not model cache or other memory effects. For object-oriented languages, our results demonstrate that an older-first collector, which collects older objects before the youngest ones, copies on average much less data than generational collectors. Our resul...

Abstract not found

. Logic programming languages are becoming more complex with the introduction of new features such as constraints or terms with an equality theory. With this increase in complexity, they require more and more sophisticated memory management. This survey gives an insight into the memory management problems in sequential logic programming language implementations; it also describes the presently known solutions. It is meant to be understood by non-specialists in logic programming with good knowledge of memory management in general. We first describe a "usefulness logic" for run-time objects. Usefulness logic defines non-garbage objects. Next, memory management systems are presented from the most trivial original run-time system, with no real concern for memory problems, to elaborated run-time systems with memory management closely observing the usefulness logic. Finally, the choice of a garbage collection technique is discussed in relation with logic programming specificities. 1 Introduc...

A garbage collection algorithm for a WAM-based Prolog system is presented. It uses a classical mark-and-compact approach, but requires less passes through the data areas than previous algorithms. It is applicable when the smallest garbage-collectable entity occupies an amount of space which is sucient to store two pointers plus 2 bits with each pointer. This collector has been implemented in the SEPIA Prolog system. 1 Introduction In state-of-the-art LISP systems, the method of choice for garbage collection is copying, which has superseded the mark-and-compact approach. Unfortunately, copying collectors usually change the order of memory cells which makes them dicult to use for Prolog. The main resons are: The Prolog terms are allocated in a stack like fashion in order to be able to pop this stack on failure (This space reclamation on failure makes it even possible to write memory-intensive Prolog programs that don't need garbage collection at all). In WAM-based Prolog implemen...

We address memory management aspects of WAM-based logic programming systems that support tabled evaluation through the use of a suspension/resumption mechanism. We describe the memory organization and usefulness logic of such systems, and issues that have to be resolved for effective and efficient garbage collection. Special attention is given to early reset in the context of suspended computations and to what is involved in the implementation of a segment order preserving copying collector for a tabled Prolog system. We also report our experience from building two working heap garbage collectors (one mark&slide and one mark&copy) for the XSB system on top of a CHAT-based tabled abstract machine: we discuss general implementation choices for heap garbage collection in `plain' WAM and issues that are specific to WAM with tabling. Finally, we compare the performance of our collectors with those of other Prolog systems and extensively analyze their characteristics. Thus, this article documents our own implementation and serves as guidance to anyone interested in proper memory management of tabled systems or systems that are similar in certain aspects.

Prolog is a logic programming language accepting a more general clause form than standard Prolog (namely hereditary Harrop formulas instead of Horn formulas) and using simply typed λ-terms as a term domain instead of first order terms. Despite these extensions, it is still amenable to goal-directed proofs and can still be given procedural semantics. However, the execution of λProlog programs requires several departures from the standard resolution scheme. First, the augmented clause form causes the program (a set of clauses) and the signature (a set of constants) to be changeable, but in a very disciplined way. Second, the new term domain has a semi-decidable and infinitary unification theory, and it introduces the need for a fi-reduction operation at run-time. MALI is an abstract memory that is suitable for storing the search-state of depth-first search processes. Its main feature is its efficient memory management. We have used an original Prolog-to-C translation: predicates are trans...

We give a logical reconstruction of all-solution predicates in terms of list comprehensions in Prolog's and we describe a variety of logic programming constructs in terms of monads and monad morphisms. Novel monad structures are described for lazy function lists, clause unfoldings and a monad morphism based embedding of Prolog in Prolog is given. Keywords: computing paradigms, logic programming, monads, list comprehensions, all solution predicates, Prolog, higher-order unification, lazy function lists. 1 Introduction Monads and comprehensions, have been successfully used in functional programming as a convenient generalization of various structurally similar programming constructs starting with simple ones like list processing and ending with intricate ones like CPS transformations and state transformers. We refer to the work of Philip Wadler [16, 17] for a long list of powerful examples and to Moggi [10] for the categorist sources of the concept of monad. All-solution predic...