Nuclear fusion is the process that powers the Sun and has the potential to generate plentiful and sustainable energy for the future.
The decision to construct the first power-plant scale fusion reactor, ITER, in Cadarache, France has dramatically increased the need to solve the remaining technical challenges of the reactor design. The reactor is planned to be operational within approximately a decade (2019) and the budget is counted in tens of billion euros. Before ITER is operational, numerical simulations are the only possible way to investigate the behaviour of full scale reactors.
The CRESTA consortium members CSC, ABO and UEDIN are partners in the EUFORIA project (EU Fusion fOR Iter Applications) and consequently have substantial experience of the performance of the major European fusion plasma codes on top-level computer systems. The experience from EUFORIA is that a small number of the current dominant fusion plasma codes have the potential to scale to 100-petaflops/s or exascale and thereby simulate the ITER reactor in much greater detail than has been achieved so far. However, none of the codes currently have this capacity. ELMFIRE is one of the codes with exascale potential. This is a first principle plasma turbulence simulation code with full-function gyrokinetics. ELMFIRE is particularly well suited for investigating edge localized modes and interactions between the plasma and the reactor wall, which contain, perhaps, the most crucial dynamics with regard to the success of ITER.
Understanding plasma turbulence is very important for the success of ITER. Understanding the physics underlying the formation and development of plasma turbulence is a key issue for predicting plasma confinement and performance for future fusion devices. Of particular importance in the study of plasma confinement is the transition between low and high confinement states, still not explained by theory although experimentally observed in 1982.
