Relatore : Jean-Luc LAMOTTE

Co-relazione : FORTIN Pierre

Contributions of hybrid architectures to depth imaging: a CPU, APU and GPU comparative study

In an exploration context, Oil and Gas (O&G) companies rely on HPC to accelerate depth imaging algorithms.
Solutions based on CPU clusters and hardware accelerators are widely embraced by the industry. The Graphics Processing Units (GPUs), with a huge compute power and a high memory bandwidth, had attracted significant interest.
However, deploying heavy imaging workflows, the Reverse Time Migration (RTM) being the most famous, on such hardware had suffered from few limitations.
Namely, the lack of memory capacity, frequent CPU-GPU communications that may be bottlenecked by the PCI transfer rate, and high power consumptions. Recently, AMD has launched the Accelerated Processing Unit (APU): a processor that merges a CPU and a GPU on the same die, with promising features notably a unified CPU-GPU memory.
Throughout this thesis, we explore how efficiently may the APU technology be applicable in an O&G context, and study if it can overcome the limitations that characterize the CPU and GPU based solutions. The APU is evaluated with the help of memory, applicative and power efficiency OpenCL benchmarks.
The feasibility of the hybrid utilization of the APUs is surveyed. The efficiency of a directive based approach is also investigated. By means of a thorough review of a selection of seismic applications (modeling and RTM) on the node level and on the large scale level, a comparative study between the CPU, the APU and the GPU is conducted. We show the relevance of overlapping I/O and MPI communications with computations for the APU and GPU clusters, that APUs deliver performances that range between those of CPUs and those of GPUs, and that the APU can be as power efficient as the GPU.

Pubblicazioni 2012-2018

• 2018
  • I. Said, P. Fortin, J.‑L. Lamotte, H. Calandra : “Leveraging the accelerated processing units for seismic imaging: A performance and power efficiency comparison against CPUs and GPUs”, International Journal of High Performance Computing Applications, vol. 32 (6), pp. 819-837, (SAGE Publications) (2018)
• 2016
  • F. Jézéquel, J.‑L. Lamotte, I. Said : “Estimation de la reproductibilité numérique dans les environnements hybrides CPU-GPU”, CANUM, mini-symposium ``Recherche reproductible'', Obernai, France (2016)
  • I. Said, P. Fortin, J.‑L. Lamotte, H. Calandra : “hiCL: an OpenCL abstraction layer for scientific computing, application to depth imaging on GPU and APU”, The ACM International Conference Series (ACM ICPS)., Vienne, Austria (2016)
  • I. Said, P. Fortin, J.‑L. Lamotte, R. Dolbeau, H. Calandra : “On the efficiency of the Accelerated Processing Unit for scientific computing”, Proceedings of the 2016 Spring Simulation Multi-Conference (SPRINGSIM)., Pasadena, United States, pp. 349-356, (Society for Computer Simulation International) (2016)
  • I. Said, P. Fortin, J.‑L. Lamotte, H. Calandra : “Efficient Reverse Time Migration on APU clusters”, 2016 Rice Oil & Gas HPC Conference, Houston, United States (2016)
• 2015
  • I. Said : “Contributions of hybrid architectures to depth imaging: a CPU, APU and GPU comparative study”, these, difesa 12/21/2015, relatore Lamotte, Jean-Luc, co-relazione : Fortin, Pierre (2015)
  • F. Jézéquel, J.‑L. Lamotte, I. Said : “Estimation of numerical reproducibility on CPU and GPU”, 8<sup>th</sup> Workshop on Computer Aspects of Numerical Algorithms (CANA), Federated Conference on Computer Science and Information Systems (FedCSIS), Lodz, Poland, pp. 687-692 (2015)
• 2014
  • P. Eberhart, I. Said, P. Fortin, H. Calandra : “Hybrid strategy for stencil computations on the APU”, Proceedings of the 1<sup>st</sup> International Workshop on High-Performance Stencil Computations, Vienna, Austria, pp. 43-49 (2014)
• 2013
  • F. Jézéquel, J.‑L. Lamotte, I. Said : “Validation numérique de codes scientifiques sur GPU”, Congrès SMAI, Biennale Française des Mathématiques Appliquées et Industrielles, mini-symposium ``Etapes vers la reproductibilité numérique des calculs'', Seignosse, France, pp. 178-179 (2013)
  • H. Calandra, R. Dolbeau, P. Fortin, J.‑L. Lamotte, I. Said : “Evaluation of Successive CPUs/APUs/GPUs Based on an OpenCL Finite Difference Stencil”, 21^st Euromicro International Conference Parallel, Distributed and Network-Based Processing, PDP 2013, Belfast, United Kingdom, pp. 405-409, (IEEE) (2013)
• 2012
  • H. Calandra, R. Dolbeau, P. Fortin, J.‑L. Lamotte, I. Said : “Assessing the relevance of APU for high performance scientific computing”, AMD Fusion Developer Summit (AFDS), Bellevue, WA, United States (2012)