GPU computing in discrete optimization. Part II: Survey focused on routing problems

Schulz, Christian; Hasle, Geir; Brodtkorb, André R.; Hagen, Trond Runar

doi:10.1007/s13676-013-0026-0

Cited by 32 publications

(15 citation statements)

References 84 publications

(146 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also discuss general development and optimization strategies for writing algorithms that may approach peak performance. In Part II [22], we give a literature survey focusing on the use of GPUs for routing problems.…”

Section: Discussionmentioning

confidence: 99%

“…For illustrative purposes we provide a profiling of our local search example in Section 7, followed by a short summary in Section 8. In Part II [22], we give a survey of the literature on GPU based methods in discrete optimization, with focus on routing problems.…”

Section: Introductionmentioning

confidence: 99%

“…The latter fact illustrates also the limited usefulness of our GPU vs. CPU speedup statement in Section 5. For a more detailed discussion about usefulness and limits of such GPU vs. CPU speedup measurements see Section 4 in the second part of this paper [22].…”

mentioning

confidence: 99%

See 2 more Smart Citations

GPU computing in discrete optimization. Part I: Introduction to the GPU

Brodtkorb

Hagen

Schulz

et al. 2013

EURO Journal on Transportation and Logistics

Self Cite

View full text Add to dashboard Cite

In many cases there is still a large gap between the performance of current optimization technology and the requirements of real world applications. As in the past, performance will improve through a combination of more powerful solution methods and a general performance increase of computers. These factors are not independent. Due to physical limits, hardware development no longer results in higher speed for sequential algorithms, but rather in increased parallelism. Modern commodity PCs include a multi-core CPU and at least one GPU, providing a low cost, easily accessible heterogeneous environment for high performance computing. New solution methods that combine task parallelization and stream processing are needed to fully exploit modern computer architectures and profit from future hardware developments. This paper is the first part of a series of two, where the goal of this first part is to give a tutorial style introduction to modern PC architectures and GPU programming. We start with a short historical account of modern mainstream computer architectures, and a brief description of parallel computing. This is followed by the evolution of modern GPUs, before a GPU programming example is given. Strategies and guidelines for program development are also discussed. Part II gives a broad survey of the existing literature on parallel computing targeted at modern PCs in discrete optimization, with special focus on papers on routing problems. We conclude with lessons learnt, directions for future research, and prospects.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GPU computing in discrete optimization. Part I: Introduction to the GPU

Brodtkorb

Hagen

Schulz

et al. 2013

EURO Journal on Transportation and Logistics

Self Cite

View full text Add to dashboard Cite

show abstract

“…While the development efforts involved in task parallelization of sequential VRP solver software may be low or moderate, the cost of developing software that also exploits data parallel accelerators such as the GPU is considerably higher. For a tutorial and literature survey focused on routing problems, we refer the reader to Brodtkorb et al [8] and Schulz et al [58].…”

Section: New and Emerging Technologiesmentioning

confidence: 99%

Chapter 12: Software Tools and Emerging Technologies for Vehicle Routing and Intermodal Transportation

Bräysy¹,

Hasle²

2014

Vehicle Routing

Self Cite

View full text Add to dashboard Cite

Vehicle routing is a central task in a large number of private and public corporations (see, e.g., Golden, Raghavan, and Wasil [30]). Tours have to be planned in very diverse sectors of the economy, not only in the logistics and transport business but in virtually all industrial sectors producing physical goods. Variants of the vehicle routing problem manifest in a remarkably wide range of commercial and non-commercial enterprises: from waste/refuse collection to retail distribution; from construction material delivery planning to postal and express delivery routing; from inbound manufacturing component transportation to finished car distribution; from in-home primary health care service to hospital operations; from transportation network optimization to third Party Logistics (3PL) operation scheduling; and from bulk collection and delivery planning to passenger transportation routing. Hence, vehicle routing is key to logistics efficiency in industry, the public sector, and society in general. The high complexity of the vehicle routing problem renders purely human planning inadequate for most applications. Typically, humanmade plans have a large potential for improvement. Also, non-assisted planning occupies valuable human resources. Therefore, high-quality software tools for decision support in vehicle routing are crucial to effective and efficient planning in many sectors of society.Vehicle Routing Problems (VRPs) in their many guises have been the subject of intensive study for more than half a century now (see Laporte [41]). These research efforts have led to the development of a variety of models and solution algorithms and the publication of thousands of scientific papers. Our ability to solve VRPs has increased tremendously over the past half century due to a combination of better VRP methods and the general performance improvement of computers. These results have, together with the progress of key enabling technologies, led to the emergence of numerous software companies worldwide, selling commercial vehicle routing software tools and accompanying services (see Hallamäki et al. [32], Drexl [13], and Partyka and Hall [50]). Among the 351 Downloaded 12/15/14 to 132.236.27.111. Redistribution subject to SIAM license or copyright; see http://www.siam.org/journals/ojsa.php

show abstract

“…This includes self-operating such systems or accessing the resources in the cloud. Only initial results are available for solving scheduling problems or vehicle routing problems based on metaheuristics using CUDA (cf., for example, Schulz et al 2013). …”

Section: Problems Concerning the Reduction Of The Model Building And mentioning

confidence: 99%