770 resultados para parallel programs
Resumo:
We propose an analysis for detecting procedures and goals that are deterministic (i.e. that produce at most one solution), or predicates whose clause tests are mutually exclusive (which implies that at most one of their clauses will succeed) even if they are not deterministic (because they cali other predicates that can produce more than one solution). Applications of such determinacy information include detecting programming errors, performing certain high-level program transformations for improving search efñciency, optimizing low level code generation and parallel execution, and estimating tighter upper bounds on the computational costs of goals and data sizes, which can be used for program debugging, resource consumption and granularity control, etc. We have implemented the analysis and integrated it in the CiaoPP system, which also infers automatically the mode and type information that our analysis takes as input. Experiments performed on this implementation show that the analysis is fairly accurate and efncient.
Resumo:
It is generally recognized that information about the runtime cost of computations can be useful for a variety of applications, including program transformation, granularity control during parallel execution, and query optimization in deductive databases. Most of the work to date on compile-time cost estimation of logic programs has focused on the estimation of upper bounds on costs. However, in many applications, such as parallel implementations on distributed-memory machines, one would prefer to work with lower bounds instead. The problem with estimating lower bounds is that in general, it is necessary to account for the possibility of failure of head unification, leading to a trivial lower bound of 0. In this paper, we show how, given type and mode information about procedures in a logic program, it is possible to (semi-automatically) derive nontrivial lower bounds on their computational costs. We also discuss the cost analysis for the special and frequent case of divide-and-conquer programs and show how —as a pragmatic short-term solution —it may be possible to obtain useful results simply by identifying and treating divide-and-conquer programs specially.
Resumo:
We provide a method whereby, given mode and (upper approximation) type information, we can detect procedures and goals that can be guaranteed to not fail (i.e., to produce at least one solution or not termínate). The technique is based on an intuitively very simple notion, that of a (set of) tests "covering" the type of a set of variables. We show that the problem of determining a covering is undecidable in general, and give decidability and complexity results for the Herbrand and linear arithmetic constraint systems. We give sound algorithms for determining covering that are precise and efiicient in practice. Based on this information, we show how to identify goals and procedures that can be guaranteed to not fail at runtime. Applications of such non-failure information include programming error detection, program transiormations and parallel execution optimization, avoiding speculative parallelism and estimating lower bounds on the computational costs of goals, which can be used for granularity control. Finally, we report on an implementation of our method and show that better results are obtained than with previously proposed approaches.
Resumo:
We present a technique to estimate accurate speedups for parallel logic programs with relative independence from characteristics of a given implementation or underlying parallel hardware. The proposed technique is based on gathering accurate data describing one execution at run-time, which is fed to a simulator. Alternative schedulings are then simulated and estimates computed for the corresponding speedups. A tool implementing the aforementioned techniques is presented, and its predictions are compared to the performance of real systems, showing good correlation.
Resumo:
This paper presents a study of the effectiveness of three different algorithms for the parallelization of logic programs based on compile-time detection of independence among goals. The algorithms are embedded in a complete parallelizing compiler, which incorporates different abstract interpretation-based program analyses. The complete system shows the task of automatic program parallelization to be practical. The trade-offs involved in using each of the algorithms in this task are studied experimentally, weaknesses of these identified, and possible improvements discussed.
Resumo:
Several types of parallelism can be exploited in logic programs while preserving correctness and efficiency, i.e. ensuring that the parallel execution obtains the same results as the sequential one and the amount of work performed is not greater. However, such results do not take into account a number of overheads which appear in practice, such as process creation and scheduling, which can induce a slow-down, or, at least, limit speedup, if they are not controlled in some way. This paper describes a methodology whereby the granularity of parallel tasks, i.e. the work available under them, is efficiently estimated and used to limit parallelism so that the effect of such overheads is controlled. The run-time overhead associated with the approach is usually quite small, since as much work is done at compile time as possible. Also, a number of run-time optimizations are proposed. Moreover, a static analysis of the overhead associated with the granularity control process is performed in order to decide its convenience. The performance improvements resulting from the incorporation of grain size control are shown to be quite good, specially for systems with médium to large parallel execution overheads.
Resumo:
We argüe that in order to exploit both Independent And- and Or-parallelism in Prolog programs there is advantage in recomputing some of the independent goals, as opposed to all their solutions being reused. We present an abstract model, called the Composition- Tree, for representing and-or parallelism in Prolog Programs. The Composition-tree closely mirrors sequential Prolog execution by recomputing some independent goals rather than fully re-using them. We also outline two environment representation techniques for And-Or parallel execution of full Prolog based on the Composition-tree model abstraction. We argüe that these techniques have advantages over earlier proposals for exploiting and-or parallelism in Prolog.
Resumo:
While logic programming languages offer a great deal of scope for parallelism, there is usually some overhead associated with the execution of goals in parallel because of the work involved in task creation and scheduling. In practice, therefore, the "granularity" of a goal, i.e. an estimate of the work available under it, should be taken into account when deciding whether or not to execute a goal concurrently as a sepárate task. This paper describes a method for estimating the granularity of a goal at compile time. The runtime overhead associated with our approach is usually quite small, and the performance improvements resulting from the incorporation of grainsize control can be quite good. This is shown by means of experimental results.
Resumo:
There has been significant interest in parallel execution models for logic programs which exploit Independent And-Parallelism (IAP). In these models, it is necessary to determine which goals are independent and therefore eligible for parallel execution and which goals have to wait for which others during execution. Although this can be done at run-time, it can imply a very heavy overhead. In this paper, we present three algorithms for automatic compiletime parallelization of logic programs using IAP. This is done by converting a clause into a graph-based computational form and then transforming this graph into linear expressions based on &-Prolog, a language for IAP. We also present an algorithm which, given a clause, determines if there is any loss of parallelism due to linearization, for the case in which only unconditional parallelism is desired. Finally, the performance of these annotation algorithms is discussed for some benchmark programs.
Resumo:
This paper presents and proves some fundamental results for independent and-parallelism (IAP). First, the paper treats the issues of correctness and efficiency: after defining strict and non-strict goal independence, it is proved that if strictly independent goals are executed in parallel the solutions obtained are the same as those produced by standard sequential execution. It is also shown that, in the absence of failure, the parallel proof procedure doesn't genérate any additional work (with respect to standard SLDresolution) while the actual execution time is reduced. The same results hold even if non-strictly independent goals are executed in parallel, provided a trivial rewriting of such goals is performed. In addition, and most importantly, treats the issue of compile-time generation of IAP by proposing conditions, to be written at compile-time, to efficiently check strict and non-strict goal independence at run-time and proving the sufficiency of such conditions. It is also shown how simpler conditions can be constructed if some information regarding the binding context of the goals to be executed in parallel is available to the compiler trough either local or program-level analysis. These results therefore provide a formal basis for the automatic compile-time generation of IAP. As a corollary of such results, the paper also proves that negative goals are always non-strictly independent, and that goals which share a first occurrence of an existential variable are never independent.
Resumo:
Effective static analyses have been proposed which infer bounds on the number of resolutions or reductions. These have the advantage of being independent from the platform on which the programs are executed and have been shown to be useful in a number of applications, such as granularity control in parallel execution. On the other hand, in distributed computation scenarios where platforms with different capabilities come into play, it is necessary to express costs in metrics that include the characteristics of the platform. In particular, it is specially interesting to be able to infer upper and lower bounds on actual execution times. With this objective in mind, we propose an approach which combines compile-time analysis for cost bounds with a one-time profiling of the platform in order to determine the valúes of certain parameters for a given platform. These parameters calíbrate a cost model which, from then on, is able to compute statically time bound functions for procedures and to predict with a significant degree of accuracy the execution times of such procedures in the given platform. The approach has been implemented and integrated in the CiaoPP system.
Resumo:
Abstract is not available.
Resumo:
We present new algorithms which perform automatic parallelization via source-to-source transformations. The objective is to exploit goal-level, unrestricted independent andparallelism. The proposed algorithms use as targets new parallel execution primitives which are simpler and more flexible than the well-known &/2 parallel operator, which makes it possible to generate better parallel expressions by exposing more potential parallelism among the literals of a clause than is possible with &/2. The main differences between the algorithms stem from whether the order of the solutions obtained is preserved or not, and on the use of determinacy information. We briefly describe the environment where the algorithms have been implemented and the runtime platform in which the parallelized programs are executed. We also report on an evaluation of an implementation of our approach. We compare the performance obtained to that of previous annotation algorithms and show that relevant improvements can be obtained.
Resumo:
Effective static analyses have been proposed which allow inferring functions which bound the number of resolutions or reductions. These have the advantage of being independent from the platform on which the programs are executed and such bounds have been shown useful in a number of applications, such as granularity control in parallel execution. On the other hand, in certain distributed computation scenarios where different platforms come into play, with each platform having different capabilities, it is more interesting to express costs in metrics that include the characteristics of the platform. In particular, it is specially interesting to be able to infer upper and lower bounds on actual execution time. With this objective in mind, we propose a method which allows inferring upper and lower bounds on the execution times of procedures of a program in a given execution platform. The approach combines compile-time cost bounds analysis with a one-time profiling of the platform in order to determine the values of certain constants for that platform. These constants calibrate a cost model which from then on is able to compute statically time bound functions for procedures and to predict with a significant degree of accuracy the execution times of such procedures in the given platform. The approach has been implemented and integrated in the CiaoPP system.
Resumo:
We propose an abstract syntax for Prolog that will help the manipulation of programs at compile-time, as well as the exchange of sources and information among the tools designed for this manipulation. This includes analysers, partial evaluators, and program transformation tools. We have chosen to concentrate on the information exchange format, rather than on the syntax of programs, for which we assume a simplified format. Our purpose is to provide a low-level meeting point for the tools which will allow them to read the same programs and understand the information about them. This report describes our first design in an informal way. We expect this design to evolve and concretize, along with the future development of the tools, during the project.