A sparse knapsack algo-tech-cuit and its synthesis

نویسندگان

  • Rumen Andonov
  • Sanjay V. Rajopadhye
چکیده

Architectures parallèles, bases de données, réseaux et systèmes distribués Abstract: We systematically derive an improved algorithm (called the sparse algorithm) for the general knapsack problem which has better average case performance than the standard (dense) dynamic programming algorithm. The derivation is based on transformation of the standard recurrences into a stream functional programs, and cannot be achieved by the usual space-time mapping techniques because the dependencies are statically unpredictable. Furthermore such a sparse algorithm for the general knapsack problem has not been proposed in the literature, to the best of our knowledge. We also implement the sparse algorithm on a linear asynchronous array with constant size memory on each PE (i.e., a wavefront array processor). The algorithm can run on any number of processors and has optimal time speedup. The second (backtracking) phase of the dynamic programming is also modiied so that it can be performed in sparsely and in a memory eecient manner. Un algorithme \creux" pour le probl eme du sac a dos: synth ese et r ealisation R esum e : Nous montrons comment d eriver syst ematiquement un algorithme am e-lior e (appel e algorithme creux) pour le probl eme g en eral du sac-a-dos avec des performances meilleures en moyenne que l'algorithme standard par programmation dynamique. Cette d erivation, qui repose sur une traduction des equations r ecur-rentes standard vers un langage fonctionnel de listes, ne peut faire appel aux techniques classiques de projection spatiale et temporelle car les d ependances ne sont pas connues statiquement. Un tel algorithme creux pour le probl eme du sac a dos g en eral n'appara^ t pas a notre conna^ ssance dans la litt erature. Nous impl ementons aussi cet algorithme sur un r eseau lin eaire asynchrone (un r eseau a fronts d'ondes) dans lequel chaque processeur el ementaire dispose d'une m emoire de taille xe. Cet algorithme creux fonctionne avec un nombre quelconque de processeurs et oore un speedup optimal. La seconde phase (le retour arri ere) de la programmation dyna-mique est egalement modii e pour s'ex ecuter aussi de mani ere creuse et econome en m emoire.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An Optimal Algo-Tech-Cuit for the Knapsack Problem

We present a formal derivation and proof of correctness of a systolic array for the knapsack problem, a well known, NP-complete problem. The dependency graph of the algorithm is not completely known statically, so the derivation also serves as a case study for systolic synthesis for this class of programs. The array is itself important since it achieves optimal performance on a model much weake...

متن کامل

A generalized implicit enumeration algorithm for a class of integer nonlinear programming problems

Presented here is a generalization of the implicit enumeration algorithm that can be applied when the objec-tive function is being maximized and can be rewritten as the difference of two non-decreasing functions. Also developed is a computational algorithm, named linear speedup, to use whatever explicit linear constraints are present to speedup the search for a solution. The method is easy to u...

متن کامل

Combinatorial Auctions, Knapsack Problems, and Hill-Climbing Search

This paper examines the performance of hill climbing algo rithms on standard test problems for combinatorial auctions CAs On single unit CAs deterministic hill climbers are found to perform well and their performance can be improved signi cantly by randomizing them and restarting them several times or by using them collectively For some problems this good performance is shown to be no better th...

متن کامل

Numerical Turbulence Simulation on di erentParallel Computers using the Sparse

The parallel numerical solution of the Navier-Stokes equations with the sparse grid combination method was studied. This algo-rithmic concept is based on the independent solution of many problems with reduced size and their linear combination. The algorithm for 3-dimensional problems is described. Its parallel implementation on an IBM SP2 and a cluster of 16 HP workstations is discussed.

متن کامل

An Efficient Algorithm for Reducing the Duality Gap in a Special Class of the Knapsack Problem

A special class of the knapsack problem is called the separable nonlinear knapsack problem. This problem has received considerable attention recently because of its numerous applications. Dynamic programming is one of the basic approaches for solving this problem. Unfortunately, the size of state-pace will dramatically increase and cause the dimensionality problem. In this paper, an efficient a...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1994