scholarly journals The 5 minute rule for trading memory for disc accesses and the 10 byte rule for trading memory for CPU time

1987 ◽  
Author(s):  
Jim Gray ◽  
Franco Putzolu
Keyword(s):  
Cpu Time

scholarly journals Discrete Vortex Prediction of Flows around a Cylinder Near a Wall Using Overlapping Grid System

Fluids ◽  
2021 ◽  
Vol 6 (6) ◽  
pp. 211
Author(s):  
Wisnu Wardhana ◽  
Ede Mehta Wardhana ◽  
Meitha Soetardjo
Keyword(s):  
Grid Point ◽  
The Body ◽  
Cylinder Surface ◽  
Grid System ◽  
Vortex Interaction ◽  
Discrete Vortex ◽  
Cpu Time ◽  
Comparison Of The Results ◽  
Circular Grid ◽  
Interaction Velocity

Modelling of unidirectional and oscillatory flows around a cylinder near a wall using an overlapping grid system is carried out. The circular grid system of the cylinder was overlapped with the rectangular grid system of the wall. The use of such an overlapping grid system is intended to reduce the CPU time compared to the cloud scheme in which vortex-to-vortex interaction is used, i.e., especially in calculating the shedding vortex velocity, since calculating the vortices velocity takes the longest CPU time. This method is not only time efficient, but also gives a better distribution of surface vorticity as the scattered vortices around the body are now concentrated on a grid point. Therefore, grid-to-grid interaction is used instead of vortex-to-vortex interaction. Velocity calculation was also carried out using this overlapping grid in which the new incremental shift position was summed up to obtain the total new vortices position. The engineering applications of this topic are to simulate the loading of submarine pipeline placed close to the seabed or to simulate the flow as a result of the scouring process below the cylinder since there is space for the fluid to flow beneath it. The in-line and transverse force coefficients are found by integrating the pressure around the cylinder surface. The flow patterns are then obtained and presented. The comparison of the results with experimental evidence is presented and the range of good results is discussed.

scholarly journals A Note on Algorithms for Genotype and Allele Elimination in Complex Pedigrees With Incomplete Genotype Data

Genetics ◽  
2000 ◽  
Vol 156 (4) ◽  
pp. 2051-2062
Author(s):  
F-X Du ◽  
I Hoeschele
Keyword(s):  
Markov Chains ◽  
Computational Efficiency ◽  
Genotype Data ◽  
Genetic Analyses ◽  
Cpu Time ◽  
Elimination Algorithm

Abstract Elimination of genotypes or alleles for each individual or meiosis, which are inconsistent with observed genotypes, is a component of various genetic analyses of complex pedigrees. Computational efficiency of the elimination algorithm is critical in some applications such as genotype sampling via descent graph Markov chains. We present an allele elimination algorithm and two genotype elimination algorithms for complex pedigrees with incomplete genotype data. We modify all three algorithms to incorporate inheritance restrictions imposed by a complete or incomplete descent graph such that every inconsistent complete descent graph is detected in any pedigree, and every inconsistent incomplete descent graph is detected in any pedigree without loops with the genotype elimination algorithms. Allele elimination requires less CPU time and memory, but does not always eliminate all inconsistent alleles, even in pedigrees without loops. The first genotype algorithm produces genotype lists for each individual, which are identical to those obtained from the Lange-Goradia algorithm, but exploits the half-sib structure of some populations and reduces CPU time. The second genotype elimination algorithm deletes more inconsistent genotypes in pedigrees with loops and detects more illegal, incomplete descent graphs in such pedigrees.

scholarly journals Fast lightweight accurate xenograft sorting

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Jens Zentgraf ◽  
Sven Rahmann
Keyword(s):  
State Of The Art ◽  
Hash Table ◽  
Human Tumor ◽  
Surrounding Tissue ◽  
Cpu Time ◽  
Alignment Free ◽  
Time Usage ◽  
The One ◽  
Similar Accuracy ◽  
Software Prefetching

Abstract Motivation With an increasing number of patient-derived xenograft (PDX) models being created and subsequently sequenced to study tumor heterogeneity and to guide therapy decisions, there is a similarly increasing need for methods to separate reads originating from the graft (human) tumor and reads originating from the host species’ (mouse) surrounding tissue. Two kinds of methods are in use: On the one hand, alignment-based tools require that reads are mapped and aligned (by an external mapper/aligner) to the host and graft genomes separately first; the tool itself then processes the resulting alignments and quality metrics (typically BAM files) to assign each read or read pair. On the other hand, alignment-free tools work directly on the raw read data (typically FASTQ files). Recent studies compare different approaches and tools, with varying results. Results We show that alignment-free methods for xenograft sorting are superior concerning CPU time usage and equivalent in accuracy. We improve upon the state of the art sorting by presenting a fast lightweight approach based on three-way bucketed quotiented Cuckoo hashing. Our hash table requires memory comparable to an FM index typically used for read alignment and less than other alignment-free approaches. It allows extremely fast lookups and uses less CPU time than other alignment-free methods and alignment-based methods at similar accuracy. Several engineering steps (e.g., shortcuts for unsuccessful lookups, software prefetching) improve the performance even further. Availability Our software xengsort is available under the MIT license at http://gitlab.com/genomeinformatics/xengsort. It is written in numba-compiled Python and comes with sample Snakemake workflows for hash table construction and dataset processing.

Tooth Thickness Effect on the Performance of Gas Labyrinth Seals

1992 ◽  
Vol 114 (4) ◽  
pp. 790-795 ◽  
Author(s):  
D. L. Rhode ◽  
R. I. Hibbs
Keyword(s):  
Finite Difference ◽  
Computer Code ◽  
Thickness Effect ◽  
Turbulence Energy ◽  
Length Scale ◽  
Labyrinth Seals ◽  
Cpu Time ◽  
Reservoir Conditions ◽  
Turbulence Length ◽  
Turbulence Length Scale

A previously validated finite difference computer code was revised to allow the specification of upstream and downstream reservoir conditions as boundary conditions, whereas the domain extends only from the seal inlet to outlet plane. As a result of this special revision, the required execution CPU time is approximately only one hour on a VAX 8650 computer for three-cavity, straight-through seals. A parametric study focusing on tooth thickness showed that streamwise swirl development was only slightly higher for the thickest tooth. Further, for straight-through seals it was found that leakage is almost independent of tooth thickness and that the second cavity yields a definite increase in turbulence energy and turbulence length scale over the first cavity.

scholarly journals Comparison of Two Numerical Algorithms for Computing the Effects of Mistuning of Fan Flutter

2016 ◽  
Author(s):  
L. Salles ◽  
M. Vahdati
Keyword(s):  
Numerical Models ◽  
Three Dimensional ◽  
Numerical Algorithms ◽  
Mode Shapes ◽  
Minimal Amount ◽  
Aeroelastic System ◽  
Cpu Time ◽  
Influence Coefficients ◽  
Non Linear ◽  
The Time Domain

The aim of this paper is to study the effects of mistuning on fan flutter and to compare the prediction of two numerical models of different fidelity. The high fidelity model used here is a three-dimensional, whole assembly, time-accurate, viscous, finite-volume compressible flow solver. The Code used for this purpose is AU3D, written in Imperial College and validated for flutter computations over many years. To the best knowledge of authors, this is the first time such computations have been attempted. This is due to the fact that, such non-linear aeroelastic computations with mistuning require large amount of CPU time and cannot be performed routinely and consequently, faster (low fidelity) models are required for this task. Therefore, the second model used here is the aeroelastic fundamental mistuning model (FMM) and it based on an eigenvalue analysis of the linearized modal aeroelastic system with the aerodynamic matrix calculated from the aerodynamic influence coefficients. The influence coefficients required for this algorithm are obtained from the time domain non-linear Code by shaking one blade in the datum (tuned) frequency and mode. Once the influence coefficients have been obtained, the computations of aero damping require minimal amount of CPU time and many different mistuning patterns can be studied. The objectives of this work are to: 1. Compare the results between the two models and establish the capabilities/limitations of aeroelastic FMM, 2. Check if the introduction of mistuning would bring the experimental and computed flutter boundaries closer, 3. Establish a relationship between mistuning and damping. A rig wide-chord fan blade, typical of modern civil designs, was used as the benchmark geometry for this study. All the flutter analyses carried out in this paper are with frequency mistuning, but the possible consequences of mistuned mode shapes are briefly discussed at the end of this paper. Only the first family of modes (1F, first flap) is considered in this work. For the frequency mistuning analysis, the 1F frequency is varied around the annulus but the 1F mode shapes remain the same for all the blades. For the mode shape mistuning computations, an FE analysis of the whole assembly different mass blades is performed. The results of this work clearly show the importance of mistuning on flutter. It also demonstrates that when using rig test data for aeroelastic validation of CFD codes, the amount mistuning present must be known. Finally, it should be noted that the aim of this paper is the study of mistuning and not steady/unsteady validation of a CFD code and therefore minimal aerodynamic data are presented.

Scalability studies and large grid computations for surface combatant using CFDShip-Iowa

2011 ◽  
Vol 25 (4) ◽  
pp. 466-487 ◽  
Author(s):  
Shanti Bhushan ◽  
Pablo Carrica ◽  
Jianming Yang ◽  
Frederick Stern
Keyword(s):  
Free Surface ◽  
Message Passing Interface ◽  
Free Surface Flows ◽  
Processing Unit ◽  
Mean Flow ◽  
Time Step ◽  
Vortical Structures ◽  
Cpu Time ◽  
Surface Combatant ◽  
Straight Ahead

Scalability studies and computations using the largest grids to date for free-surface flows are performed using message-passing interface (MPI)-based CFDShip-Iowa toolbox curvilinear (V4) and Cartesian (V6) grid solvers on Navy high-performance computing systems. Both solvers show good strong scalability up to 2048 processors, with V6 showing somewhat better performance than V4. V6 also outperforms V4 in terms of the memory requirements and central processing unit (CPU) time per time-step per grid point. The explicit solvers show better scalability than the implicit solvers, but the latter allows larger time-step sizes, resulting in a lower total CPU time. The multi-grid HYPRE solver shows better scalability than the portable, extensible toolkit for scientific computation solver. The main scalability bottleneck is identified to be the pressure Poisson solver. The memory bandwidth test suggests that further scalability improvements could be obtained by using hybrid MPI/open multi-processing (OpenMP) parallelization. V4-detached eddy simulation (DES) on a 300 M grid for the surface combatant model DTMB 5415 in the straight-ahead condition provides a plausible description of the vortical structures and mean flow patterns observed in the experiments. However, the vortex strengths are over predicted and the turbulence is not resolved. V4-DESs on up to 250 M grids for DTMB 5415 at 20° static drift angle significantly improve the forces and moment predictions compared to the coarse grid unsteady Reynolds averaged Navier–Stokes, due to the improved resolved turbulence predictions. The simulations provide detailed resolution of the free-surface and breaking pattern and vortical and turbulent structures, which will guide planned experiments. V6 simulations on up to 276 M grids for DTMB 5415 in the straight-ahead condition predict diffused vortical structures due to poor wall-layer predictions. This could be due to the limitations of the wall-function implementation for the immersed boundary method.

On the choice of CFD codes in the design process of planing sailing yachts

2009 ◽  
Author(s):  
Jérémie Raymond ◽  
Jean-Marie Finot ◽  
Jean-Michel Kobus ◽  
Gérard Delhommeau ◽  
Patrick Queutey ◽  
...  
Keyword(s):  
Free Surface ◽  
Design Process ◽  
Finite Differences ◽  
Potential Flow ◽  
Surface Condition ◽  
Cpu Time ◽  
Free Surface Condition ◽  
Finite Differences Method ◽  
Non Linear ◽  
Ranse Solver

The discussion is based on results gathered during the first two years of a 3 years research program for the benefits of Groupe Finot-Conq, Naval Architects. The introduction presents the objectives of the program: Setting up a practical method using numerical and experimental available tools to design fast planing sailing yachts. The aim of this paper is to compare advantages and disadvantages of four different kinds of CFD codes which are linear and non-linear potential flow approach, RANSE solver using finite differences method and RANSE solver using volume of fluid method. The Fluid Mechanics Laboratory of the Ecole Centrale de Nantes (France) has developed those three approaches so those homemade codes will be used for this study. The first one is REVA, a potential flow code with a linearised free surface condition. ICARE is a RANSE solver using finite differences method with a non linear free surface condition. It is extensively used for industrial projects as for sailing yachts projects (ACC for example). ISIS-CFD is a RANSE solver using finite volume method to build the spatial discretization of the transport equations with unstructured mesh. The latter is able to compute sprays for fast planing ships but is also the slower in terms of CPU time. In addition, we had the opportunity to test FS-FLOW which is a potential flow code with a non linear free surface condition distributed by FRIENDSHIP CONSULTING. Numerical results for the four codes are compared with the other codes' results as with tank tests data. Those tank tests were made using captive model test technique on two Open60' models. Reasons of the choice of the captive model technique are explained and experimental procedures are briefly described. Comparisons between codes are mainly based on the easiness of use, the cost in CPU time and the confidence we can have in the results as a function of the boat speed. Flow visualizations, pressure maps, free surface deformation are shown and compared. Analysis of local quantities integrated or by zone is also presented. Results are analyzed focusing on the ability of each code to represent flow dynamics for every speed with a special attention to high speeds. The practical question raised is to know which kind of answers each code can bring in terms of tendencies evaluation or sensitivity to hull geometry modifications. The main goal is to be able to judge if those codes are able to make reliable and consistent comparisons of different designs. Conclusion is that none of the codes is perfect and gather all the advantages. It is still difficult to propose a definitive methodology to estimate hydrodynamic performances at every speed and at every stage of the design process. Knowing each code limitations, it appears more coherent to use each of them at different stages of the design process: the quickest and less reliable to understand the main tendencies and the longest and more precise to validate the final options.

Numerical seismograms obtained using ω-κ integrals, for a point P source in a vertically inhomogeneous anelastic model

1996 ◽  
Vol 86 (3) ◽  
pp. 750-760
Author(s):  
F. Abramovici ◽  
L. H. T. Le ◽  
E. R. Kanasewich
Keyword(s):  
Half Space ◽  
Homogeneous Layer ◽  
Variable Step Size ◽  
Step Size ◽  
Adaptive Process ◽  
Cpu Time ◽  
Homogeneous Half Space ◽  
Linear Layer ◽  
Variable Step ◽  
Viscoelastic Layers

Abstract This article presents some numerical experiments in using a computer program for calculating the displacements due to a P source in a vertically inhomogeneous structure, based on the Fourier-Bessel representation. The structure may contain homogeneous, inhomogeneous, elastic, or viscoelastic layers. The source may act in any type of sublayer or in the half-space. Synthetic results for the simple case of a homogeneous layer overlaying a homogeneous half-space compare favorably with computations based on the Cagniard method. Numerical seismograms for an elastic layer having velocities and density varying linearly with depth were computed by integrating numerically the governing differential systems and compared with results based on the Haskell model of splitting the linear layer in homogeneous sublayers. Even an adaptive process with a variable step size based on the Haskell model has a poorer performance on the accuracy-cpu time scale than numerical integration.

Sign in / Sign up

Export Citation Format

Share Document