Hybrid MPI/OpenMP cloud parallelization of harmonic coupled finite strip method applied on reinforced concrete prismatic shell structure

Nikolić, Milutin; Hajdukovic, Miroslav; Milašinović, Dragan D.; Goleš, Danica; Maric, Petar; Živanov, Žarko

doi:10.1016/j.advengsoft.2014.12.006

Cited by 10 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It was inevitable to combine different parallelization models -MPI and OpenMP and use different sources of parallel resources (multicomputer, multiprocessormulticore, GPGPU based, or cloud available environment). Our results show that rational usage of hybrid parallelization approach to speed up the critical parts of our code contributes in significant performance improvements [9,10]. If uniformly compressed thin-walled structures undergo inelastic deformations, these structures generally sustain two sources of non-linearity (geometrical non-linearity due to large deflection and material non-linearity due to inelastic behaviour).…”

Section: Faculty Of Civil Engineering Subotica University Of Novi Samentioning

confidence: 88%

See 1 more Smart Citation

Influence of Mode Interactions on Inelastic Buckling Effective Stresses in Thin-Walled Columns Using Finite Strip Method

Milašinović¹,

Maric²,

Živanov³

et al. 2019

Zbornik GFS

Self Cite

View full text Add to dashboard Cite

The inelastic buckling, free vibration and damage of prismatic thin-walled columns, considering two sources of non-linearity (geometrical non-linearity due to large deflection and material non-linearity due to inelastic behaviour), are analysed by implementing the semi-analytical finite strip method (FSM). A very basic continuum damage model with one damage parameter is implemented in conjunction with a mathematical material modelling approach named rheological-dynamical analogy (RDA) in order to address stiffness reduction due to inelastic behaviour. The RDA is a type of inelastic analysis, which transforms one category of very complicated material non-linear problems to simpler linear dynamical problems using modal analysis. The effective stress, as defined in the damage mechanics, is used in this paper. An analysis of damaged composite thin-walled wide-flange columns made of a polymer resin, which is reinforced with fibrous material, is carried out. Numerical examples computed by application of extensive hybrid parallelization present the influence of mode interactions on the damage variables and effective stresses in thin-walled columns.

show abstract

Section: Faculty Of Civil Engineering Subotica University Of Novi Samentioning

confidence: 88%

“…8, is analysed taking into account the governing dynamic RDA modulus along the longitudinal axis. The HCFSM algorithm for geometric non-linear analysis has been described in [9] and [10]. The semi-analytical FSM eigenvalue analysis of a structure is simpler as it is explained in Section 4.1.…”

Section: Numerical Applicationsmentioning

confidence: 99%

Influence of Mode Interactions on Inelastic Buckling Effective Stresses in Thin-Walled Columns Using Finite Strip Method

Milašinović¹,

Maric²,

Živanov³

et al. 2019

Zbornik GFS

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recent works have been proposed to port scientific applications from various domains to private and public clouds [54][55][56][57] by using a mixture of MPI and OpenMP; an approach that benefits from the communication efficiency of MPI primitives and the easy parallelization of OpenMP annotations. Experimental results usually show good performance, but they also reveal the difficulties associated to MPI programming, which requires a level of expertise and platform knowledge that is far beyond the knowledge of typical programmers.…”

Section: Related Workmentioning

confidence: 99%

Cluster Programming using the OpenMP Accelerator Model

Yviquel

Cruz

Araújo

2018

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Computation offloading is a programming model in which program fragments (e.g., hot loops) are annotated so that their execution is performed in dedicated hardware or accelerator devices. Although offloading has been extensively used to move computation to GPUs, through directive-based annotation standards like OpenMP, offloading computation to very large computer clusters can become a complex and cumbersome task. It typically requires mixing programming models (e.g., OpenMP and MPI) and languages (e.g., C/C++ and Scala), dealing with various access control mechanisms from different cloud providers (e.g., AWS and Azure), and integrating all this into a single application. This article introduces computer cluster nodes as simple OpenMP offloading devices that can be used either from a local computer or from the cluster head-node. It proposes a methodology that transforms OpenMP directives to Spark runtime calls with fully integrated communication management, in a way that a cluster appears to the programmer as yet another accelerator device. Experiments using LLVM 3.8, OpenMP 4.5 on well known cloud infrastructures (Microsoft Azure and Amazon EC2) show the viability of the proposed approach, enable a thorough analysis of its performance, and make a comparison with an MPI implementation. The results show that although data transfers can impose overheads, cloud offloading from a local machine can still achieve promising speedups for larger granularity: up to 115× in 256 cores for the 2MM benchmark using 1GB sparse matrices. In addition, the parallel implementation of a complex and relevant scientific application reveals a 80× speedup on a 320 core machine when executed directly from the headnode of the cluster. CCS Concepts: • Computing methodologies → Distributed programming languages; • Software and its engineering → Parallel programming languages; Distributed programming languages; Runtime environments; Source code generation;

show abstract

“…Recent works have been proposed to port scientific applications from various domains to private and public clouds [36], [37], [38], [39] by using a mixture of MPI and OpenMP; an approach that benefits from the communication efficiency of MPI primitives and the easy parallelization of OpenMP annotations. Experimental results usually show good performance, but they also reveal the difficulties associated to MPI programming, which require a level of expertise and platform knowledge that is far beyond the knowledge of typical programmers.…”

Section: Related Workmentioning

confidence: 99%

The Cloud as an OpenMP Offloading Device

Yviquel¹,

Araújo²

2017

2017 46th International Conference on Parallel Processing (ICPP)

View full text Add to dashboard Cite

Abstract-Computation offloading is a programming model in which program fragments (e.g. hot loops) are annotated so that their execution is performed in dedicated hardware or accelerator devices. Although offloading has been extensively used to move computation to GPUs, through directive-based annotation standards like OpenMP, offloading computation to very large computer clusters can become a complex and cumbersome task. It typically requires mixing programming models (e.g. OpenMP and MPI) and languages (e.g. C/C++ and Scala), dealing with various access control mechanisms from different clouds (e.g. AWS and Azure), and integrating all this into a single application. This paper introduces the cloud as a computation offloading device. It integrates OpenMP directives, cloud based mapreduce Spark nodes and remote communication management such that the cloud appears to the programmer as yet another device available in its local computer. Experiments using LLVM, OpenMP 4.5 and Amazon EC2 show the viability of the proposed approach and enable a thorough analysis of the performance and costs involved in cloud offloading. The results show that although data transfers can impose overheads, cloud offloading can still achieve promising speedups of up to 86x in 256 cores for the 2MM benchmark using 1GB matrices.

show abstract

Hybrid MPI/OpenMP cloud parallelization of harmonic coupled finite strip method applied on reinforced concrete prismatic shell structure

Cited by 10 publications

References 7 publications

Influence of Mode Interactions on Inelastic Buckling Effective Stresses in Thin-Walled Columns Using Finite Strip Method

Influence of Mode Interactions on Inelastic Buckling Effective Stresses in Thin-Walled Columns Using Finite Strip Method

Cluster Programming using the OpenMP Accelerator Model

The Cloud as an OpenMP Offloading Device

Contact Info

Product

Resources

About