Cactus Framework: Black Holes to Gamma Ray Bursts

Petascale computing will require additional ingredients over conventional parallelism. We consider some of the challenges which will be caused by future petascale architectures, and discuss our plans for the future development of the Cactus framework and its applications to meet these challenges in order to profit from these new architectures.

Ninety years after Einstein first proposed his General Theory of Relativity (GR), astrophysicists are more than ever and in greater detail probing into regions of the universe where gravity is very strong and where, according to GR’s geometric description, the curvature of spacetime is large.

The realm of strong curvature is notoriously difficult to investigate with conventional observational astronomy, and some phenomena might bear no observable electro-magnetic signature at all and may only be visible in neutrinos (if sufficiently close to Earth) or in gravitational wavesripples of spacetime itself which are predicted by Einstein’s GR. Gravitational waves have not been observed directly to date, but gravitational-wave detectors (e.g., LIGO [1], GEO [2], VIRGO [3]) are in the process of reaching sensitivities sufficiently high to observe interesting astrophysical phenomena.

Until gravitational-wave astronomy becomes reality, astrophysicists must rely on computationally and conceptually challenging large-scale numerical simulations in order to grasp the details of the energetic processes occurring in regions of strong curvature that are shrouded from direct observation in the electromagnetic spectrum by intervening matter or that have little or no electromagnetic signature at all. Such astrophysical systems and phenomena include the birth of neutron FIG. 1: Left: Gravitational waves and horizons in a binary black hole inspiral simulation. Simulation by AEI/CCT collaboration, image by W. Benger (CCT/AEI/ZIB). Right: Rotationally deformed protoneutron star formed in the iron core collapse of an evolved massive star. Shown are a volume rendering of the restmass density and a 2D rendition of outgoing gravitational waves. Simulation by [4], image by R. Kähler. stars (NSs) or black holes (BHs) in collapsing evolved massive stars, coalescence of compact1 binary systems, gamma-ray bursts (GRBs), active galactic nuclei harboring supermassive black holes, pulsars, and quasi-periodically oscillating NSs (QPOs). In figure 1 we present example visualizations of binary BH and stellar collapse calculations carried out by our groups.

From these, GRBs, intense narrowly-beamed flashes of γ-rays of cosmological origin, are among the most scientifically interesting and the riddle concerning their central engines and emission mechanisms is one of the most complex and challenging problems of astrophysics today.

GRBs last between 0.5-1000 secs, with a bimodal distribution of durations [5], indicating two distinct classes of mechanisms and central engines. The short-hard (duration < ∼ 2 secs) group of GRBs (hard, because their γ-ray spectra peak at shorter wavelength) predominantly occurs in elliptical galaxies with old stellar populations at moderate astronomical distances [5,6]. The energy released in a short-hard GRB and its duration suggest [5,6] a black hole with a ∼0.1 solar-mass (M ⊙ ) accretion disk as the central engine. Such a BH-accretion-disk system is likely to be formed by the coalescence of NS-NS or NS-BH systems (e.g., [7]).

Long-soft (duration ∼2-1000 secs) GRBs on the other hand seem to occur exclusively in the starforming regions of spiral or irregular galaxies with young stellar populations and low metallicity2 Observations that have recently become available (see [5] for reviews) indicate features in the x-ray and optical afterglow spectra and luminosity evolutions of long-soft GRBs that show similarities with spectra and light curves obtained from Type-Ib/c core-collapse supernovae whose progenitors are evolved massive stars (M > ∼ 25 M ⊙ ) that have lost their extended hydrogen envelopes and probably also a fair fraction of their helium shell. These observations support the collapsar model [6] of long-soft GRBs that envisions a stellar-mass black hole formed in the aftermath of a stellar corecollapse event with a massive ∼1 M ⊙ rotationally-supported accretion disk as the central engine, powering the GRB jet that punches through the compact and rotationally-evacuated polar stellar envelope reaching ultr

…(Full text truncated)…