Nektar++ v5.0.0 released
The latest version of Nektar++, v5.0.0, was released on the 9th December 2019. It can be downloaded from the downloads page.
Spectral/hp element methods for turbomachinery
A T106A low pressure turbine vane was investigated at moderate regime (Re = 88450), exploring the convergence properties of the main flow statistics with the aim of developing a set of best practices to achieve DNS resolution.
Nektar++ Workshop 2019
The 4th annual Nektar++ workshop will be held this year between June 10-12th 2019 at the University of Exeter, UK. For full details, see the main Nektar++ workshop page.
Clowder The Clowder cluster is currently composed by 3 Odroid-C2 single board computers (SBC), a 5 port Gigabit Ethernet switch (Netgear GS105) and a 5V 12A power supply (see Figure 1). The Odroid-C2 is a SBC equipped with a Amlogic ARM Cortex-A53(ARMv8) 1.5Ghz quad core CPU, 2 Gbyte DDR3 SDRAM, and Gigabit Ethernet. The Odroid-C2 […]
Building Nektar++ on platforms without outgoing Internet access
This blog post explains how to manually work around this issue by placing the correct third party dependencies in the correct location on the target system such that Nektar++ is able to discover and use them during the build process without needing to attempt to download the files.
Nektar++ and hydrodynamic stability in undulated channels
Nektar++ has been used to study various forms of hydrodynamic instabilities arising is a flow through a channel with corrugated walls. Stability analysis has been performed using direct numerical simulation and tracking growth, or attenuation of the unstable modes.
Dispersion/diffusion analysis and LES
Nektar++ has been recently used to verify the dispersion/diffusion estimates of discontinuous spectral element methods for non-periodic flow problems. This work a key implications on the effective construction of LES tools for industrial problems.
Nektar++ on the UCL Thomas cluster
Brief description of and Nektar++ installation instructions for Thomas, the UK National Tier 2 High Performance Computing Hub in Materials and Molecular Modelling.
Nektar++ strong scaling performance on Mira
We have recently performed some tests of the Nektar++ incompressible Navier-Stokes solver on a Formula 1 car geometry. The tests were run on Mira up to 131k cores and show excellent scaling of of the code for massively parallel simulations.
A Programme for the Nektar++ workshop is now available: Nektar++ Workshop 2017