-
Towards Confidential Computing: A Secure Cloud Architecture for Big Data Analytics and AI
Authors:
Naweiluo Zhou,
Florent Dufour,
Vinzent Bode,
Peter Zinterhof,
Nicolay J Hammer,
Dieter Kranzlmüller
Abstract:
Cloud computing provisions computer resources at a cost-effective way based on demand. Therefore it has become a viable solution for big data analytics and artificial intelligence which have been widely adopted in various domain science. Data security in certain fields such as biomedical research remains a major concern when moving their workflows to cloud, because cloud environments are generally…
▽ More
Cloud computing provisions computer resources at a cost-effective way based on demand. Therefore it has become a viable solution for big data analytics and artificial intelligence which have been widely adopted in various domain science. Data security in certain fields such as biomedical research remains a major concern when moving their workflows to cloud, because cloud environments are generally outsourced which are more exposed to risks. We present a secure cloud architecture and describes how it enables workflow packaging and scheduling while keeping its data, logic and computation secure in transit, in use and at rest.
△ Less
Submitted 28 May, 2023;
originally announced May 2023.
-
Exploiting the Space Filling Curve Ordering of Particles in the Neighbour Search of Gadget3
Authors:
Antonio Ragagnin,
Nikola Tchipev,
Michael Bader,
Klaus Dolag,
Nicolay J. Hammer
Abstract:
Gadget3 is nowadays one of the most frequently used high performing parallel codes for cosmological hydrodynamical simulations. Recent analyses have shown t\ hat the Neighbour Search process of Gadget3 is one of the most time-consuming parts. Thus, a considerable speedup can be expected from improvements of the u\ nderlying algorithms. In this work we propose a novel approach for speeding up the N…
▽ More
Gadget3 is nowadays one of the most frequently used high performing parallel codes for cosmological hydrodynamical simulations. Recent analyses have shown t\ hat the Neighbour Search process of Gadget3 is one of the most time-consuming parts. Thus, a considerable speedup can be expected from improvements of the u\ nderlying algorithms. In this work we propose a novel approach for speeding up the Neighbour Search which takes advantage of the space-filling-curve particle ordering. Instead of performing Neighbour Search for all particles individually, nearby active particles can be grouped and one single Neighbour Search can be performed to obta\ in a common superset of neighbours. Thus, with this approach we reduce the number of searches. On the other hand, tree walks are performed within a larger searching radius. There is an optimal size of grouping that maximize the speedup, which we found by numerical experiments. We tested the algorithm within the boxes of the Magneticum project. As a result we obtained a speedup of $1.65$ in the Density and of $1.30$ in the Hydrodynamics computation, respectively, and a total speedup of $1.34.$
△ Less
Submitted 23 October, 2018;
originally announced October 2018.
-
A web portal for hydrodynamical, cosmological simulations
Authors:
Antonio Ragagnin,
Klaus Dolag,
Veronica Biffi,
Marionne Cadolle Bel,
Nicolay J. Hammer,
Aliaksei Krukau,
Margarita Petkova,
Daniel Steinborn
Abstract:
This article describes a data center hosting a web portal for accessing and sharing the output of large, cosmological, hydro-dynamical simulations with a broad scientific community. It also allows users to receive related scientific data products by directly processing the raw simulation data on a remote computing cluster. The data center has a multi-layer structure: a web portal, a job control la…
▽ More
This article describes a data center hosting a web portal for accessing and sharing the output of large, cosmological, hydro-dynamical simulations with a broad scientific community. It also allows users to receive related scientific data products by directly processing the raw simulation data on a remote computing cluster. The data center has a multi-layer structure: a web portal, a job control layer, a computing cluster and a HPC storage system. The outer layer enables users to choose an object from the simulations. Objects can be selected by visually inspecting 2D maps of the simulation data, by performing highly compounded and elaborated queries or graphically by plotting arbitrary combinations of properties. The user can run analysis tools on a chosen object. These services allow users to run analysis tools on the raw simulation data. The job control layer is responsible for handling and performing the analysis jobs, which are executed on a computing cluster. The innermost layer is formed by a HPC storage system which hosts the large, raw simulation data. The following services are available for the users: (I) {\sc ClusterInspect} visualizes properties of member galaxies of a selected galaxy cluster; (II) {\sc SimCut} returns the raw data of a sub-volume around a selected object from a simulation, containing all the original, hydro-dynamical quantities; (III) {\sc Smac} creates idealised 2D maps of various, physical quantities and observables of a selected object; (IV) {\sc Phox} generates virtual X-ray observations with specifications of various current and upcoming instruments.
△ Less
Submitted 19 October, 2018; v1 submitted 19 December, 2016;
originally announced December 2016.
-
Performance Optimisation of Smoothed Particle Hydrodynamics Algorithms for Multi/Many-Core Architectures
Authors:
Fabio Baruffa,
Luigi Iapichino,
Nicolay J. Hammer,
Vasileios Karakasis
Abstract:
We describe a strategy for code modernisation of Gadget, a widely used community code for computational astrophysics. The focus of this work is on node-level performance optimisation, targeting current multi/many-core IntelR architectures. We identify and isolate a sample code kernel, which is representative of a typical Smoothed Particle Hydrodynamics (SPH) algorithm. The code modifications inclu…
▽ More
We describe a strategy for code modernisation of Gadget, a widely used community code for computational astrophysics. The focus of this work is on node-level performance optimisation, targeting current multi/many-core IntelR architectures. We identify and isolate a sample code kernel, which is representative of a typical Smoothed Particle Hydrodynamics (SPH) algorithm. The code modifications include threading parallelism optimisation, change of the data layout into Structure of Arrays (SoA), auto-vectorisation and algorithmic improvements in the particle sorting. We obtain shorter execution time and improved threading scalability both on Intel XeonR ($2.6 \times$ on Ivy Bridge) and Xeon PhiTM ($13.7 \times$ on Knights Corner) systems. First few tests of the optimised code result in $19.1 \times$ faster execution on second generation Xeon Phi (Knights Landing), thus demonstrating the portability of the devised optimisation solutions to upcoming architectures.
△ Less
Submitted 10 May, 2017; v1 submitted 19 December, 2016;
originally announced December 2016.
-
The world's largest turbulence simulations
Authors:
Christoph Federrath,
Ralf S. Klessen,
Luigi Iapichino,
Nicolay J. Hammer
Abstract:
Understanding turbulence is critical for a wide range of terrestrial and astrophysical applications. Here we present first results of the world's highest-resolution simulation of turbulence ever done. The current simulation has a grid resolution of 10048^3 points and was performed on 65536 compute cores on SuperMUC at the Leibniz Supercomputing Centre (LRZ). We present a scaling test of our modifi…
▽ More
Understanding turbulence is critical for a wide range of terrestrial and astrophysical applications. Here we present first results of the world's highest-resolution simulation of turbulence ever done. The current simulation has a grid resolution of 10048^3 points and was performed on 65536 compute cores on SuperMUC at the Leibniz Supercomputing Centre (LRZ). We present a scaling test of our modified version of the FLASH code, which updates the hydrodynamical equations in less than 3 micro seconds per cell per time step. A first look at the column density structure of the 10048^3 simulation is presented and a detailed analysis is provided in a forthcoming paper.
△ Less
Submitted 3 July, 2016;
originally announced July 2016.
-
An axis-free overset grid in spherical polar coordinates for simulating 3D self-gravitating flows
Authors:
Annop Wongwathanarat,
Nicolay J. Hammer,
Ewald Müller
Abstract:
A type of overlapping grid in spherical coordinates called the Yin-Yang grid is successfully implemented into a 3D version of the explicit Eulerian grid-based code PROMETHEUS including self-gravity. The modified code successfully passed several standard hydrodynamic tests producing results which are in very good agreement with analytic solutions. Moreover, the solutions obtained with the Yin-Yang…
▽ More
A type of overlapping grid in spherical coordinates called the Yin-Yang grid is successfully implemented into a 3D version of the explicit Eulerian grid-based code PROMETHEUS including self-gravity. The modified code successfully passed several standard hydrodynamic tests producing results which are in very good agreement with analytic solutions. Moreover, the solutions obtained with the Yin-Yang grid exhibit no peculiar behaviour at the boundary between the two grid patches. The code has also been successfully used to model astrophysically relevant situations, namely equilibrium polytropes, a Taylor-Sedov explosion, and Rayleigh-Taylor instabilities. According to our results, the usage of the Yin-Yang grid greatly enhances the suitability and efficiency of 3D explicit Eulerian codes based on spherical polar coordinates for astrophysical flows.
△ Less
Submitted 8 March, 2010;
originally announced March 2010.
-
Three-Dimensional Simulations of Mixing Instabilities in Supernova Explosions
Authors:
N. J. Hammer,
H. -Th. Janka,
E. Mueller
Abstract:
We present the first three-dimensional (3D) simulations of the large-scale mixing that takes place in the shock-heated stellar layers ejected in the explosion of a 15.5 solar-mass blue supergiant star. The outgoing supernova shock is followed from its launch by neutrino heating until it breaks out from the stellar surface more than two hours after the core collapse. Violent convective overturn i…
▽ More
We present the first three-dimensional (3D) simulations of the large-scale mixing that takes place in the shock-heated stellar layers ejected in the explosion of a 15.5 solar-mass blue supergiant star. The outgoing supernova shock is followed from its launch by neutrino heating until it breaks out from the stellar surface more than two hours after the core collapse. Violent convective overturn in the post-shock layer causes the explosion to start with significant asphericity, which triggers the growth of Rayleigh-Taylor (RT) instabilities at the composition interfaces of the exploding star. Deep inward mixing of hydrogen (H) is found as well as fast-moving, metal-rich clumps penetrating with high velocities far into the H-envelope of the star as observed, e.g., in the case of SN 1987A. Also individual clumps containing a sizeable fraction of the ejected iron-group elements (up to several 0.001 solar masses) are obtained in some models. The metal core of the progenitor is partially turned over with Ni-dominated fingers overtaking oxygen-rich bullets and both Ni and O moving well ahead of the material from the carbon layer. Comparing with corresponding 2D (axially symmetric) calculations, we determine the growth of the RT fingers to be faster, the deceleration of the dense metal-carrying clumps in the He and H layers to be reduced, the asymptotic clump velocities in the H-shell to be higher (up to ~4500 km/s for the considered progenitor and an explosion energy of 10^{51} ergs, instead of <2000 km/s in 2D), and the outward radial mixing of heavy elements and inward mixing of hydrogen to be more efficient in 3D than in 2D. We present a simple argument that explains these results as a consequence of the different action of drag forces on moving objects in the two geometries. (abridged)
△ Less
Submitted 31 March, 2010; v1 submitted 24 August, 2009;
originally announced August 2009.
-
VLT spectroscopy and non-LTE modeling of the C/O-dominated accretion disks in two ultracompact X-ray binaries
Authors:
K. Werner,
T. Nagel,
T. Rauch,
N. J. Hammer,
S. Dreizler
Abstract:
We present new medium-resolution high-S/N optical spectra of the ultracompact low-mass X-ray binaries 4U0614+091 and 4U1626-67, taken with the ESO Very Large Telescope. They are pure emission line spectra and the lines are identified as due to C II-IV and O II-III Line identification is corroborated by first results from modeling the disk spectra with detailed non-LTE radiation transfer calculat…
▽ More
We present new medium-resolution high-S/N optical spectra of the ultracompact low-mass X-ray binaries 4U0614+091 and 4U1626-67, taken with the ESO Very Large Telescope. They are pure emission line spectra and the lines are identified as due to C II-IV and O II-III Line identification is corroborated by first results from modeling the disk spectra with detailed non-LTE radiation transfer calculations. Hydrogen and helium lines are lacking in the observed spectra. Our models confirm the deficiency of H and He in the disks. The lack of neon lines suggests an Ne abundance of less than about 10 percent (by mass), however, this result is uncertain due to possible shortcomings in the model atom. These findings suggest that the donor stars are eroded cores of C/O white dwarfs with no excessive neon overabundance. This would contradict earlier claims of Ne enrichment concluded from X-ray observations of circumbinary material, which was explained by crystallization and fractionation of the white dwarf core.
△ Less
Submitted 24 January, 2006;
originally announced January 2006.
-
On Possible Oxygen/Neon White Dwarfs: H1504+65 and the White Dwarf Donors in Ultracompact X-ray Binaries
Authors:
K. Werner,
N. J. Hammer,
T. Nagel,
T. Rauch,
S. Dreizler
Abstract:
We discuss the possibility to detect O/Ne white dwarfs by evidence for Ne overabundances. The hottest known WD, H1504+65, could be the only single WD for which we might be able to proof its O/Ne nature directly. Apart from this, strong Ne abundances are known or suspected only from binary systems, namely from a few novae, and from a handful of LMXBs. We try to verify the hypothesis that the latt…
▽ More
We discuss the possibility to detect O/Ne white dwarfs by evidence for Ne overabundances. The hottest known WD, H1504+65, could be the only single WD for which we might be able to proof its O/Ne nature directly. Apart from this, strong Ne abundances are known or suspected only from binary systems, namely from a few novae, and from a handful of LMXBs. We try to verify the hypothesis that the latter might host strongly ablated O/Ne WD donors, by abundance analyses of the accretion disks in these systems. In any case, to conclude on O/Ne WDs just by strong Ne overabundances is problematic, because Ne enrichment also occurs by settling of this species into the core of C/O WDs.
△ Less
Submitted 28 October, 2004;
originally announced October 2004.
-
Spectroscopic Analyses of the Blue Hook Stars in NGC 2808: A More Stringent Test of the Late Hot Flasher Scenario
Authors:
S. Moehler,
A. V. Sweigart,
W. B. Landsman,
N. J. Hammer,
S. Dreizler
Abstract:
Recent UV observations of the globular cluster NGC2808 (Brown et al. 2001) show a significant population of hot stars fainter than the zero-age horizontal branch (``blue hook'' stars), which cannot be explained by canonical stellar evolution. Their results suggest that stars which experience unusually large mass loss on the red giant branch and which subsequently undergo the helium core flash wh…
▽ More
Recent UV observations of the globular cluster NGC2808 (Brown et al. 2001) show a significant population of hot stars fainter than the zero-age horizontal branch (``blue hook'' stars), which cannot be explained by canonical stellar evolution. Their results suggest that stars which experience unusually large mass loss on the red giant branch and which subsequently undergo the helium core flash while descending the white dwarf cooling curve could populate this region. Theory predicts that these ``late hot flashers'' should show higher temperatures than the hottest canonical horizontal branch (HB) stars and should have He- and C-rich atmospheres. As a test of this late hot flasher scenario, we have obtained and analysed medium resolution spectra of a sample of blue hook stars in NGC2808 to derive their atmospheric parameters. Using the same procedures, we have also re-analyzed our earlier spectra of the blue hook stars in omega Cen (Moehler et al. 2002) for comparison with the present results for NGC2808. The blue hook stars in these two clusters are both hotter (Teff > 35,000 K) and more helium-rich than canonical extreme HB stars in agreement with the late hot flasher scenario. Moreover, we find indications for C enhancement in the three most He-enriched stars in NGC2808. However, the blue hook stars still show some H in their atmospheres, perhaps indicating that some residual H survives a late hot flash and then later diffuses to the surface during the HB phase. We note that the presence of blue hook stars apparently depends mostly on the total mass of the globular cluster and not so much on its HB morphology.
△ Less
Submitted 10 November, 2003;
originally announced November 2003.
