scholarly journals Memory-aware Workload Balancing Technique based on Decision Trees for Parallel HEVC Video Coding

2020 ◽  
Vol 15 (3) ◽  
pp. 1-9
Author(s):  
Iago Storch ◽  
Bruno Zatt ◽  
Luciano Agostini ◽  
Guilherme Correa ◽  
Daniel Palomino

Video coding applications demand high computational effort to achieve high compression rates at a low perceptual quality expense. In order to reach acceptable encoding time for such applications, modern video coding standards have been em-ploying parallelism approaches to exploit multiprocessing plat-forms, such as the tiling tool from HEVC standard. When employing Tiles, each frame is divided into rectangular-shaped regions which can be encoded independently. However, alt-hough it is possible to distribute the data equally among the processing units when using Tiles, balancing the workload among processing units poses great challenges. Therefore, this paper proposes a workload balancing technique aiming to speed up the HEVC parallel encoding using Tiles. Different from other literature works, the proposed solution uses a novel approach employing static uniform tiling to avoid memory management difficulties that may emerge when dynamic tiling solutions are employed. The proposed technique relies on workload distribution history of previous frames to predict the workload distribution of the current frame. Then, the pro-posed technique balances the workload among Tiles by em-ploying a workload reduction scheme based on decision trees in the coding process. Experimental tests show that the pro-posed solution outperforms the standard uniform tiling and it is competitive with related works in terms of speedup. Moreo-ver, the solution optimizes resources usage in multiprocessing platforms, presents a negligible coding efficiency loss and avoids increasing memory bandwidth usage by 9.8%, on aver-age, when compared to dynamic tiling solutions, which can impact significantly the performance in memory-constrained platforms.

Author(s):  
László Takács ◽  
Ferenc Szabó

AbstractPolymer sandwich structures have high bending stiffness and strength and also low weight. Therefore, they are widely used in the transportation industry. In the conceptual design phase, it is essential to have a method to model the mechanical behavior of the sandwich and its adhesive joints accurately in full-vehicle scale to investigate different structure partitioning strategies. In this paper, a novel approach using finite element modeling is introduced. The sandwich panels are modeled with layered shells and the joint lines with general stiffness matrices. Stiffness parameters of the face-sheets and the core material are obtained via mechanical tests. Stiffness parameters of the joints are determined by using the method of Design of Experiments, where detailed sub-models of the joints serve as a reference. These models are validated with experimental tests of glass-fiber reinforced vinyl ester matrix composite sandwich structure with a foam core. By using two joint designs and three reference geometries, it is shown that the method is suitable to describe the deformation behavior in a full-vehicle scale with sufficient accuracy.


2021 ◽  
Author(s):  
Eleonore Roguet ◽  
Emmanuel Persent ◽  
Daniel Averbuch

Abstract A new method which uses elastic and elasto-plastic Finite Element analyses is developed to design a double breech-block type connector. All relevant criteria proposed by API16F are fulfilled. In addition, plastic and bearing criteria have been added to support the use of lugs for load transfer in the connector. The proposed methodology has been applied and validated through experimental tests at different scales and in particular on laboratory specimens and small-scaled connectors. Based on these last structural tests, a safety factor of almost 8 was obtained for the design method on small-scaled connectors. Prototype tests at scale 1:1 allowed the methodology to be fully validated and a new product to be qualified. Certification bodies validated the whole design process, the employed methodology and the new connector.


2019 ◽  
Vol 21 (9) ◽  
pp. 1577-1583
Author(s):  
Carlos Guardiola ◽  
Benjamín Pla ◽  
Marcelo Real ◽  
Cyril Travaillard ◽  
Frederic Dambricourt

The impact of short-circuit pulses on the after-treatment system of a spark-ignited engine must be taken into account to keep the fuel-to-air equivalence ratio within the three-way catalyst window, thereby reducing pollutant emissions. The fuel-to-air equivalence ratio overestimation that suffers the wide-range λ-sensor upstream three-way catalyst in the presence of short circuit is especially relevant. In this study, a novel approach to deal with the fuel-to-air equivalence ratio control under short-circuit conditions is introduced. Under this scope, this work proposes a strategy for the on-board correction of the aforementioned fuel-to-air equivalence ratio overestimation, by means of the information regarding short-circuit level that provides the frequency content of the λ-sensor at the engine frequency. Finally, the potential of this approach to minimize pollutant emissions, in particular the NO x penalty arisen as a consequence of running the engine under leaner conditions than expected, is assessed through experimental tests.


2018 ◽  
Vol 2018 ◽  
pp. 1-14
Author(s):  
A. Y. Elruby ◽  
Sam Nakhla ◽  
A. Hussein

The eXtended Finite Element Method (XFEM) is a versatile method for solving crack propagation problems. Meanwhile, XFEM predictions for crack onset and propagation rely on the stress field which tends to converge at a slower rate than that of displacements, making it challenging to capture critical load at crack onset accurately. Furthermore, identifying the critical region(s) for XFEM nodal enrichments is user-dependent. The identification process can be straightforward for small-scale test specimen while in other cases such as complex structures it can be unmanageable. In this work a novel approach is proposed with three major objectives; (1) alleviate user-dependency; (2) enhance predictions accuracy; (3) minimize computational effort. An automatic critical region(s) identification based on material selected failure criterion is developed. Moreover, the approach enables the selection of optimized mesh necessary for accurate prediction of failure loads at crack initiation. Also, optimal enrichment zone size determination is automated. The proposed approach was developed as an iterative algorithm and implemented in ABAQUS using Python scripting. The proposed algorithm was validated against our test data of unnotched specimens and relevant test data from the literature. The results of the predicted loads/displacements at failure are in excellent agreement with measurements. Crack onset locations were in very good agreement with observations from testing. Finally, the proposed algorithm has shown a significant enhancement in the overall computational efficiency compared to the conventional XFEM. The proposed algorithm can be easily implemented into user-built or commercial finite element codes.


2013 ◽  
Vol 816-817 ◽  
pp. 1169-1173
Author(s):  
Usman Attique ◽  
Abdul Ghafoor ◽  
Riaz Ahmed ◽  
Shahid Ikramullah

Various exact and heuristic methods have been proposed for assembly line balancing problem (ALBP) but unequal multiple operators have not been considered much. In this paper we present a novel approach of assembly line balancing Type-2 with unequal multiple operators by using an already developed code in Matlab (Tomlab modeling platform). The adopted approach can be applied at line balancing problems ranging from few to hundreds of work elements to achieve minimum cycle time with very less computational effort.


Author(s):  
Fabrizio Pistani ◽  
Daniel Brooker

The design approach for the analysis of a riser turret mooring system (RTM) connected to a FPSO is presented here. The design was carried out with the use of numerical models developed with Ariane and Orcaflex software packages. Ariane has been used as a screening tool while the model developed with Orcaflex was used for detailed analysis. The succesful validation of the models against existing experimental tests helps to avoid the repetition of a model test campaign for design of new systems. An original solution to overcome the limitations of the mooring software Ariane has been developed and used for mooring analysis. This paper describes the development of the numerical models highlighting the original features, the validation with the model tests and discusses some of the main results.


Geophysics ◽  
1996 ◽  
Vol 61 (3) ◽  
pp. 666-682 ◽  
Author(s):  
Michael S. Zhdanov ◽  
Peter Traynin ◽  
John R. Booker

A new method of the resistivity imaging based on frequency‐domain electromagnetic migration is developed. Electromagnetic (EM) migration involves downward diffusion of observed EM fields whose time flow has been reversed. Unlike downward analytical continuation, migration is a stable procedure that accurately restores the phase of the upgoing field inside the Earth. This method is indented for the processing and interpretation of EM data collected for both TE and TM modes of plane‐wave excitation. Until recently, the method could be applied only for determining the position of anomalous structures and for finding interfaces between layers of different conductivity. There were no well developed approaches to the resistivity imaging, which is the key problem in the inversion of EM data. We provide a novel approach to determining not only the position of anomalous structures but their resistivity as well. The main difficulty in the practical realization of this approach is determining the background resistivity distribution for migration. We discuss the method of the solution of this problem based on differential transformation of apparent resistivity curves. The final goal of migration is to provide a first order interpretation using a computational effort equivalent to a forward modeling calculation.


Sign in / Sign up

Export Citation Format

Share Document