New GPGPU technologies, such as CUDA Dynamic Parallelism (CDP), can help dealing with recursive patterns of computation, such as divide-and-conquer, used by backtracking algorithms. The initial part of this thesis proposes a GPU-accelerated backtracking algorithm using CDP that extends a well-known parallel backtracking model. Unlike related works from the literature, the proposed algorithm does not dynamically allocate memory on GPU. The memory required by the subsequent kernel generations is preallocated based on the analysis of a partial backtracking tree. The second part of this work generalizes the ideas of the first algorithm for approaches that dynamically allocate memory on GPU and launch more than two kernel generations. The final ...
In the paper we present the parallel implementation of the alpha-beta algorithm running on the graph...
Combinatorial optimization problems are difficult problems whose solution by exact methods can be ti...
As Central Processing Units (CPUs) and Graphical Processing Units (GPUs) get progressively better, d...
International audienceNew GPGPU technologies, such as CUDA Dynamic Parallelism (CDP), can help deali...
International audienceCUDA Dynamic Parallelism (CDP) is an extension of the GPGPU programming model ...
International audienceThis work presents a GPU-based backtracking algorithm for permutation combinat...
In this paper we present how recent hardware revisions and newly introduced approaches to thread col...
Abstract. We explore the backtracking paradigm with properties seen as sub-optimal for GPU architect...
International audienceTree-based exploratory methods, like Branch-and-Bound (B&B) algorithms, are hi...
In this paper we present a dynamic programming based technique that is suitable for providing exact ...
Abstract—Many general-purpose applications exploit Graphics Processing Units (GPUs) by executing a s...
Propositional model counting (MC) and its extensions as well as applications in the area of probabil...
Kinematic synthesis, at its heart, involves finding the zero-dimensional solution set of asystem of p...
In the paper we present the parallel implementation of the alpha-beta algorithm running on the graph...
Combinatorial optimization problems are difficult problems whose solution by exact methods can be ti...
As Central Processing Units (CPUs) and Graphical Processing Units (GPUs) get progressively better, d...
International audienceNew GPGPU technologies, such as CUDA Dynamic Parallelism (CDP), can help deali...
International audienceCUDA Dynamic Parallelism (CDP) is an extension of the GPGPU programming model ...
International audienceThis work presents a GPU-based backtracking algorithm for permutation combinat...
In this paper we present how recent hardware revisions and newly introduced approaches to thread col...
Abstract. We explore the backtracking paradigm with properties seen as sub-optimal for GPU architect...
International audienceTree-based exploratory methods, like Branch-and-Bound (B&B) algorithms, are hi...
In this paper we present a dynamic programming based technique that is suitable for providing exact ...
Abstract—Many general-purpose applications exploit Graphics Processing Units (GPUs) by executing a s...
Propositional model counting (MC) and its extensions as well as applications in the area of probabil...
Kinematic synthesis, at its heart, involves finding the zero-dimensional solution set of asystem of p...
In the paper we present the parallel implementation of the alpha-beta algorithm running on the graph...
Combinatorial optimization problems are difficult problems whose solution by exact methods can be ti...
As Central Processing Units (CPUs) and Graphical Processing Units (GPUs) get progressively better, d...