The TORUS radiation transfer code

Harries, Tim J.; Haworth, Thomas J.; Acreman, David M.; Ali, Ahmad A.; Douglas, Tom

doi:10.1016/j.ascom.2019.03.002

Cited by 57 publications

(49 citation statements)

References 179 publications

(158 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the Monte Carlo (MC) radiative transfer (RT) and hydrodynamics (HD) code, torus (described in detail by Harries et al 2019). We use the same treatment as in Paper I, but provide a summary here.…”

Section: Methodsmentioning

confidence: 99%

Massive star feedback in clusters: variation of the FUV interstellar radiation field in time and space

Ali

Harries

2019

Monthly Notices of the Royal Astronomical Society

Self Cite

View full text Add to dashboard Cite

We investigate radiative feedback from a 34 M star in a 10 4 M turbulent cloud using three-dimensional radiation-hydrodynamics (RHD) models. We use Monte Carlo radiative transfer to accurately compute photoionization equilibrium and radiation pressure, with multiple atomic species and silicate dust grains. We include the diffuse radiation field, dust absorption/re-emission, and scattering. The cloud is efficiently dispersed, with 75 per cent of the mass leaving the (32.3 pc) 3 grid within 4.3 Myr (1.1 t ff ). This compares to all mass exiting within 1.6 Myr (0.74 t ff ) in our previously published 10 3 M cloud. At most 20 per cent of the mass is ionized, compared to 40 per cent in the lower mass model, despite the ionized volume fraction being 80 per cent in both, implying the higher mass cloud is more resilient to feedback. The total Jeans-unstable mass increases linearly up to 1500 M before plateauing after 2 Myr, corresponding to a core formation efficiency of 15 per cent. We also measure the time-variation of the far-ultraviolet (FUV) radiation field, G 0 , impinging on other cluster members, taking into account for the first time how this changes in a dynamic cluster environment with intervening opacity sources and stellar motions. Many objects remain shielded in the first 0.5 Myr whilst the massive star is embedded, after which G 0 increases by orders of magnitude. Gas motions later on cause comparable drops which happen instantaneously and last for ∼ 1 Myr before being restored. This highly variable UV field will influence the photoevaporation of protoplanetary discs near massive stars.

show abstract

Section: Methodsmentioning

confidence: 99%

Massive star feedback in clusters: variation of the FUV interstellar radiation field in time and space

Ali

Harries

2019

Monthly Notices of the Royal Astronomical Society

Self Cite

View full text Add to dashboard Cite

show abstract

“…established ray-tracing framework within the torus code (Rundle et al 2010;Harries et al 2019), with abundances and level populations fed in from the dynamical-PDR model. The only quantities that we have to make further assumptions for (i.e.…”

Section: Calculating Synthetic Alma Observationsmentioning

confidence: 99%

The observational anatomy of externally photoevaporating planet-forming discs – I. Atomic carbon

Haworth

Owen

2020

Monthly Notices of the Royal Astronomical Society

Self Cite

View full text Add to dashboard Cite

We demonstrate the utility of CI as a tracer of photoevaporative winds that are being driven from discs by their ambient UV environment. Commonly observed CO lines only trace these winds in relatively weak UV environments and are otherwise dissociated in the wind at the intermediate to high UV fields that most young stars experience. However, CI traces unsubtle kinematic signatures of a wind in intermediate UV environments (∼ 1000 G 0 ) and can be used to place constraints on the kinematics and temperature of the wind. In CI position-velocity diagrams external photoevaporation results in velocities that are faster than those from Keplerian rotation alone, as well as emission from quadrants of position-velocity space in which there would be no Keplerian emission. This is independent of viewing angle because the wind has components that are perpendicular to the azimuthal rotation of the disc. At intermediate viewing angles (∼ 30 − 60 • ) moment 1 maps also exhibit a twisted morphology over large scales (unlike other processes that result in twists, which are typically towards the inner disc). CI is readily observable with ALMA, which means that it is now possible to identify and characterise the effect of external photoevaporation on planet-forming discs in intermediate UV environments.

show abstract

“…The main bottleneck in contemporary MCRT algorithms is the grid structure that is used to discretize the interstellar medium in space. In multi-node distributed memory setups, it is easily understood that this grid structure leads to significant overheads, either because the entire structure needs to be duplicated on all nodes Camps & Baes 2020), or because photon packets need to be communicated between nodes (Harries et al 2019). On shared-memory systems, the grid is a bottleneck as well, however, this time due to bandwidth issues.…”

Section: Introductionmentioning

confidence: 99%

“…These subgrids can then act as the small memory resources on which the tasks in the taskbased algorithm operate. As in the case of a distributed-memory grid, this inevitably means storing photon packets that transition from one such subgrid to a neighbouring subgrid, which requires the use of buffers (Harries et al 2019).…”

Section: Introductionmentioning

confidence: 99%

CMACIONIZE 2.0: a novel task-based approach to Monte Carlo radiation transfer

Vandenbroucke

Camps

2020

A&A

View full text Add to dashboard Cite

Context. Monte Carlo radiative transfer (MCRT) is a widely used technique to model the interaction between radiation and a medium. It plays an important role in astrophysical modelling and when these models are compared with observations. Aims. We present a novel approach to MCRT that addresses the challenging memory-access patterns of traditional MCRT algorithms, which prevent an optimal performance of MCRT simulations on modern hardware with a complex memory architecture. Methods. We reformulated the MCRT photon-packet life cycle as a task-based algorithm, whereby the computation is broken down into small tasks that are executed concurrently. Photon packets are stored in intermediate buffers, and tasks propagate photon packets through small parts of the computational domain, moving them from one buffer to another in the process. Results. Using the implementation of the new algorithm in the photoionization MCRT code CMACIONIZE 2.0, we show that the decomposition of the MCRT grid into small parts leads to a significant performance gain during the photon-packet propagation phase, which constitutes the bulk of an MCRT algorithm because memory caches are used more efficiently. Our new algorithm is faster by a factor 2 to 4 than an equivalent traditional algorithm and shows good strong scaling up to 30 threads. We briefly discuss adjustments to our new algorithm and extensions to other astrophysical MCRT applications. Conclusions. We show that optimising the memory access patterns of a memory-bound algorithm such as MCRT can yield significant performance gains.

show abstract

The TORUS radiation transfer code

Cited by 57 publications

References 179 publications

Massive star feedback in clusters: variation of the FUV interstellar radiation field in time and space

Massive star feedback in clusters: variation of the FUV interstellar radiation field in time and space

The observational anatomy of externally photoevaporating planet-forming discs – I. Atomic carbon

CMACIONIZE 2.0: a novel task-based approach to Monte Carlo radiation transfer

Contact Info

Product

Resources

About