Online
(This seminar is organized by the IAU G5 commission on stellar and planetary atmospheres)
Task-based computing is a method where computational problems are split
into a large number of semi-independent tasks (cf.
2018MNRAS.477..624N). The method is a general one, with application not
limited to traditional grid-based simulations; it can be applied with
advantages also to particle-based and hybrid simulations, which involve
both particles and fields. The main advantages emerge when doing
simulations of very complex and / or multi-scale systems, where the
cost of updating is very unevenly distributed in space, with perhaps
large volumes with very low update cost and small but important regions
with large update costs.
Possible applications in the context of stellar atmospheres include
modelling that covers large scales, such as whole active regions on the
Sun or even the entire Sun, while at the same time allows resolving
small-scale details in the photosphere, chromosphere, and corona. In
the context of planetary atmospheres, models of pebble-accreting hot
primordial atmospheres that cover all scales, from the surfaces of
Mars- and Earth-size embryos to the scale heights of the surrounding
protoplanetary disks, have already been computed (2018MNRAS.479.5136P,
2019MNRAS.482L.107P), and one can envision a number of applications
where the task-based computing advantage is leveraged, for example to
selectively do the detailed chemistry necessary to treat atmospheres
saturated with evaporated solids, or to do complex cloud chemistry
combined with 3-D radiative transfer.
In the talk I will give a quick overview of the principles behind
task-based computing, and then use both already published and still
on-going work to illustrate how this may be used in practice. I will
finish by discussing how these methods could be applied with great
advantage to problems such as non-equilibrium ionization, non-LTE
radiative transfer, and partial redistribution diagnostics of spectral
lines.