A new workflow has been developed for predict-first equilibrium modeling. Here, the motivation is to predict DIII-D equilibria without having analyzed experimental profiles. The workflow includes a cold start tool, Modelprofiles, which constructs all the profiles and power flows needed by transport codes starting from only global quantities. MP has been coupled to the VMOMS equilibrium, to obtain an initial approximate geometry, as well as the NBEAMS pencil NBI code. The workflow also couples the EFIT code and an equilibrium validation code which computes various local and global figures of merit to assess the accuracy of a given equilibrium prediction. Using an empirical pressure model in Modelprofiles, based on the Cordey-Thomsen H-mode scaling, we find an RMS error q-profile is 21% for a database of 37 DIII-D H-mode discharges. Here, the predicted equilibrium was compared to the experimental EFIT01 magnetic reconstruction. For many of the discharges, similar results are obtained using the q-profile from EFIT02 MSE reconstructions. The automated workflow is fast with typical run times of less than 10 secs on a local workstation. Our ultimate goal is to use theory-based transport model predictions of the pressure profile in predictive EFIT calculations for between shot analysis and plasma control applications.

The newest version of the integrated modeling framework OMFIT, v0.26.0, was publicly released. The newest stable release features more than 3500 source code commits by 38 different US and international contributors. Some of the most notable improvements to the framework include an overhauled installation system, modernization of the GUI elements’ look-and-feel, and the adaptation of many key OMFIT classes to support reading/writing data to the ITER Integrated Modelling & Analysis Suite (IMAS) via the OMAS library. This latter effort compelled some developers to make the OMFIT classes COCOS compliant, which ultimately provided the foundation for adapting key modules (such as OMFITprofiles and TRANSP) to other devices, notably JET, MAST, and TCV. Of the 90+ physics modules in OMFIT, over 80 of them have been updated. OMFIT now counts over 600 unique usernames, and this year’s APS Division of Plasma Physics Meeting in Portland featured over 53 talks and posters that included some work done via OMFIT. More statistics and insights were presented at a lunchtime User Meeting at APS [https://goo.gl/oegfKZ], and useful feedback was obtained from users.

An article describing the CGYRO gyrokinetic turbulence code, which highlighted the significant performance improvements achieved through recent GPU acceleration optimizations, was published in the latest issue of HPCWire.
https://www.hpcwire.com/2018/12/05/optimizing-key-plasma-physics-code-for-latest-gen-nvidia-gpus-yields-threefold-increase-in-processing-speed/