We consider the solution of single commodity strictly convex network flow problem in a distributed asynchronous computation environment. The dual this is unconstrained, differentiable, and well suited for via Gauss–Seidel relaxation. show that structure allows successful application method whereby relaxation iterations are carried out parallel by several processors arbitrary order with arbitrarily large interprocessor communication delays.

In this paper, we propose an efficient implementation of the branch and bound method for knapsack problems on a CPU-GPU system via CUDA. Branch computations can be carried out either CPU or GPU according to size list. A better management GPUs memories, less GPU-CPU communications synchronization between threads are proposed in new order increase efficiency. Indeed, series computational results is displayed analyzed showing substantial speedup Tesla C2050 GPU.

The Simplex algorithm is a well known method to solve linear programming (LP) problems. In this paper, we propose an implementation via CUDA of the on multi GPU architecture. Computational tests have been carried out randomly generated instances for non-sparse LP show maximum speedup 24.5 with two Tesla C2050 boards.

The Simplex algorithm is a well known method to solve linear programming (LP) problems. In this paper, we propose parallel implementation of the on CPU-GPU systems via CUDA. Double precision used in order improve quality solutions. Computational tests have been carried out randomly generated instances for non-sparse LP show maximum speedup 12:5 GTX 260 board.

This paper deals with a new class of parallel asynchronous iterative algorithms for the solution nonlinear systems equations. The main feature methods presented here is possibility flexible communication between processors. In particular partial updates can be exchanged. Approximation associated fixed point mapping also considered. A detailed convergence study presented. connection Schwarz alternating method made boundary value problems. Computational results on shared memory multiprocessor...

In the last decade, Graphics Processing Units(GPUs) have gained an increasing popularity as accelerators for High Performance Computing (HPC) applications. Recent GPUs are not only powerful graphics engines but also highly threaded parallel computing processors that can achieve sustainable speedup compared with CPUs. this context, researchers try to exploit capability of architecture solve difficult problems in many domains science and engineering. article, we present recent advances on GPU...

Hybrid implementation via CUDA of a branch and bound method for knapsack problems is proposed. Branch computations can be carried out either on the CPU or GPU according to size list, i.e. number nodes. Tests are Tesla C2050 GPU. A first series computational results showing substantial speedup displayed analyzed.

This paper deals with high performance Peer-to-Peer computing applications. We concentrate on the solution of large scale numerical simulation problems via distributed iterative methods. present current version an environment that allows direct communication between peers. is based a self-adaptive protocol. The protocol configures itself automatically and dynamically in function application requirements like scheme computation elements context topology by choosing most appropriate mode A...

In this paper, we present a heuristic which derives feasible solution for the Multiple Knapsack Problem (MKP). The proposed called RCH, is recursive method that performs computation on core of knapsacks. RCH compared with MTHM Martello and Toth. Computational results randomly generated instances show approach gives better gap smaller restitution times.

The implementation via CUDA of a hybrid dense dynamic programming method for knapsack problems on amulti-GPU architecture is considered. Tests are carried out Bull cluster with Tesla S1070 computing systems. A first series computational results shows substantial speedup. speedup factor close to 28 two GPUs.

The connection between Internet of Things (IoT) and High Performance Computing (HPC) is investigated in this keynote presentation. New paradigms devices for HPC are presented. Several examples related to smart building management, logistics manufacturing leading difficult combinatorial optimization problems detailed.

Misinformation posted on social media during COVID-19 is one main example of infodemic data. This phenomenon was prominent in China when happened at the beginning. While a lot data can be collected from various platforms, publicly available detection remains rare and not easy to construct manually. Therefore, instead developing techniques for detection, this paper aims constructing Chinese dataset, “infodemic 2019”, by collecting widely spread outbreak. Each record labeled as true, false or...

This article presents an exact cooperative method for the solution of multidimensional knapsack problem (MKP) which combines dynamic programming and branch bound. Our makes cooperate a heuristics based on surrogate relaxation bound procedure. algorithm was tested several randomly generated test sets problems in literature. Solution values first step are compared with optimal results provided by other well-known existing heuristics. Then, our is classical algorithm. [Received 8 October 2008;...