Tobias Brandvik

Acceleration of a 3D Euler Solver Using Commodity Graphics Hardware

The porting of twoand three-dimensional Euler solvers from a conventional CPU implementation to the novel target platform of the Graphics Processing Unit (GPU) is described. The motivation for such an effort is the impressive performance that GPUs offer: typically 10 times more floating point operations per second than a modern CPU, with over 100 processing cores and all at a very modest financial cost. Both codes were found to generate the same results on the GPU as the FORTRAN versions did on the CPU. The 2D solver ran up to 29 times quicker on the GPU than on the CPU; the 3D solver 16 times faster. Nomenclature cv Specific heat capacity at constant volume e Specific total energy = cvT + 1 2 V 2 h0 Specific stagnation enthalpy p Pressure t Time u,v Cartesian components of velocity V Velocity (magnitude) T Temperature Yp Stagnation pressure loss coefficient = p01−p0 p01−p2 tb302@cam.ac.uk gp10006@cam.ac.uk

An Accelerated 3D Navier-Stokes Solver for Flows in Turbomachines

A new three-dimensional Navier-Stokes solver for flows in turbomachines has been developed. The new solver is based on the latest version of the Denton codes, but has been implemented to run on Graphics Processing Units (GPUs) instead of the traditional Central Processing Unit (CPU). The change in processor enables an order-of-magnitude reduction in run-time due to the higher performance of the GPU. Scaling results for a 16 node GPU cluster are also presented, showing almost linear scaling for typical turbomachinery cases. For validation purposes, a test case consisting of a three-stage turbine with complete hub and casing leakage paths is described. Good agreement is obtained with previously published experimental results. The simulation runs in less than 10 minutes on a cluster with four GPUs.Copyright

Acceleration of a two-dimensional Euler flow solver using commodity graphics hardware

Abstract The implementation of a two-dimensional Euler solver on graphics hardware is described. The graphics processing unit is highly parallelized and uses a programming model that is well suited to flow computation. Results for a transonic turbine cascade test-case are presented. For large grids (106 nodes) a 40 times speed-up compared with a Fortran implementation on a contemporary CPU is observed.

An Accelerated 3D Navier–Stokes Solver for Flows in Turbomachines

A new three-dimensional Navier–Stokes solver for flows in turbomachines has been developed. The new solver is based on the latest version of the Denton codes but has been implemented to run on graphics processing units (GPUs) instead of the traditional central processing unit. The change in processor enables an order-of-magnitude reduction in run-time due to the higher performance of the GPU. The scaling results for a 16 node GPU cluster are also presented, showing almost linear scaling for typical turbomachinery cases. For validation purposes, a test case consisting of a three-stage turbine with complete hub and casing leakage paths is described. Good agreement is obtained with previously published experimental results. The simulation runs in less than 10 min on a cluster with four GPUs.

SBLOCK: A Framework for Efficient Stencil-Based PDE Solvers on Multi-core Platforms

We present a new software framework for the implementation of applications that use stencil computations on block-structured grids to solve partial differential equations. A key feature of the framework is the extensive use of automatic source code generation which is used to achieve high performance on a range of leading multi-core processors. Results are presented for a simple model stencil running on Intel and AMD CPUs as well as the NVIDIA GT200 GPU. The generality of the frame- work is demonstrated through the implementation of a complete application consisting of many different stencil computations, taken from the field of computational fluid dynamics.

Angle-of-attack effects on counter-rotating propellers at take-off

Due to their potential for significant fuel consumption savings, Counter-Rotating Open Rotors (CRORs) are currently being considered as an alternative to high-bypass turbofans. When CRORs are mounted on an aircraft, several ‘installation effects’ arise which are not present when the engine is operated in isolation. This paper investigates how flow features arising from one such effect — the angle-of-attack of the engine centre-line relative to the oncoming flow — can influence the design of CROR engines. Three-dimensional full-annulus unsteady CFD simulations are used to predict the time-varying flow field experienced by each rotor and emphasis is put on the interaction of the front-rotor wake and tip vortex with the rear-rotor.A parametric study is presented that quantifies the rotor-rotor interaction as a function of the angle-of-attack. It is shown that angle-of-attack operation significantly changes the flow field and the unsteady lift on both rotors. In particular, a frequency analysis shows that the unsteady lift exhibits sidebands around the rotor-rotor interaction frequencies. Further, a non-linear increase in the total rear-rotor tip unsteadiness is observed for moderate and high angles-of-attack.The results presented in this paper demonstrate that common techniques used to mitigate CROR noise, such as modifying the rotor-rotor axial spacing and rear-rotor crop, can not be applied correctly unless angle-of-attack effects are taken into account.Copyright

How to improve open rotor aerodynamics at cruise and take-off

A key challenge in open rotor design is getting the optimum aerodynamics at both the cruise and take-off conditions. This is particularly difficult because the operation and the requirements of an open rotor are very different at cruise compared to takeoff. This paper uses CFD results to explore the impact of various design changes on the cruise and take-off flow-fields. The paper then considers how a given open rotor design is best operated at take-off to minimise noise whilst maintaining high thrust. The main findings are that various design modifications can be applied to control the flow features that lead to lost efficiency at cruise and increased noise emission at take-off. A breakdown of the lost power terms from CFD solutions demonstrates how developments in open rotor design have led to reduced aerodynamic losses. At take-off, the operating point of the open rotor should be set such that the non-dimensional lift is as high as possible, without causing significant flow separation. This can be achieved through suitable amounts of re-pitch and speed up applied to a design. Comparisons with fully three-dimensional CFD show that the amount of re-pitch required can be determined using simplified methods such as two-dimensional CFD and a Blade Element Method.

Large-Scale Gas Turbine Simulations on GPU Clusters

Publisher Summary This chapter presents a strategy for implementing solvers for partial differential equations (PDEs) that rely heavily on stencil computations on three-dimensional, multiblock structured grids. As a starting point, a simple stencil computation arising from the discretization of the three-dimensional heat diffusion equation is considered. Building on this example, the steps taken to redevelop a complete computational fluid dynamics solver originally written in Fortran 77 that consists of many complicated stencil computations are then described. The new solver makes extensive use of automatic source code generation for the implementation of its stencil computations. This capability is provided by a recently developed software framework called SBLOCK and a description of this framework and the strategies it uses to achieve good performance on NVIDIA graphics processing unit (GPUs) is also included. As a final demonstration of the performance and scalability of the new solver, the chapter includes both single-GPU and multi-GPU benchmarks obtained on a 64-GPU cluster.