arXiv subjects

Geray S. Karademir

Publications and source records attributed to Geray S. Karademir.

8 recordsLinked to original sources

arXiv2026

The OpenGadget3 Code for Cosmological Simulations

We present the public release of OpenGadget3 (OG3), a substantially extended and re-organised version of the widely used Gadget-2/3 family of cosmological simulation codes. Since G2 was made publicly available, the absence of continued official development led to the creation of numerous, mutually incompatible developer versions of G3 across the community. This made it effectively impossible to track bug fixes, reproduce published results, or consistently credit the many individual contributors. OG3 addresses this by consolidating these disparate developments into a single, documented, and continuously maintained code base, released under the GNU GPLv3 through a GitLab repository (with an accompanying wiki), regularly synchronised with the active developer branch, and supported by a dedicated continuous-integration (CI) pipeline. OG3 retains the core algorithmic backbone of its predecessors, a Barnes\&Hut oct-tree combined with a Particle-Mesh method for gravity, and a choice of Smoothed Particle Hydrodynamics or Meshless-Finite-Mass solvers for gas dynamics, within a hybrid MPI/OpenMP parallelisation framework, extended with GPU offloading via OpenACC and OpenMP. The code integrates an extensive suite of sub-resolution and physical modules, including primordial chemistry, radiative cooling, star formation, stellar feedback and chemical enrichment, dust formation and evolution, black hole growth and AGN feedback, MHD, thermal conduction, physical viscosity, SIDM, and massive neutrinos. OG3 introduces new features improving robustness, reproducibility, and usability: a flexible mixed-precision framework spanning 16- to 128-bit representations; expert-level tagged parameters with restart-time change log; extensive consistency checks against invalid configurations; and FAIR-compliant reporting of all settings, parameters, and specific publications. (shortened and abridged)

astro-ph.IM↗

arXiv2026

Is RISC-V Ready for Massively Parallel Astrophysical Codes?

We present a performance and portability evaluation of three well-established astrophysical production codes, namely iPIC3D, PLUTO, and OpenGadget3, on a Sophgo SG2044 RISC-V processor (part of the Monte Cimone cluster), with comparisons to AMD EPYC 9554 (x86) and NVIDIA GH200 Grace (ARM) systems. These applications represent memory-bound, compute-bound, and hybrid workloads, respectively. Numerical correctness is verified across all platforms, confirming portability. RISC-V shows consistently lower performance, with slowdowns of about $3-6\times$ relative to x86 and $5-9\times$ relative to ARM. The gap is mainly due to limited memory bandwidth, shared cache constraints, narrower 128-bit vector units, and lower clock frequency, but also less-mature auto-vectorization capability of the GNU compiler suite. Memory-bound kernels are the most affected, where early bandwidth saturation and L2 cache contention reduce scalability at higher thread counts. Hybrid MPI+OpenMP configurations reveal a trade-off between memory contention and communication overhead, with intermediate configurations achieving the best performance. These results suggest that RISC-V is capable of supporting scientific workloads; however, additional improvements in both hardware and compiler technology, particularly in auto-vectorization, are required to achieve competitive performance.

cs.DC↗

arXiv2026

SPACE-Timers -- A Stack-Based Hierarchical Timing System for C++

SPACE-Timers are a lightweight hierarchical profiling framework for C++ designed for modern high-performance computing (HPC) applications. It uses a stack-based timing model to capture deeply nested execution patterns with minimal overhead, representing runtime behaviour as a tree of timing nodes with precise attribution. The framework provides structured reports with recursive aggregation, detection of unaccounted time, and compact visual summaries of runtime distribution, supporting both quick inspection and detailed analysis. It also includes checkpointing and error detection mechanisms. SPACE-Timers supports multiple profiling backends, including NVTX, ITT, ROCtx, and Omnitrace, and integrates with the MERIC runtime system to enable energy-aware optimisation. Its successful use in OpenGadget3 demonstrates its effectiveness for large-scale scientific applications.

astro-ph.IM↗

arXiv2025

EuroHPC SPACE CoE: Redesigning Scalable Parallel Astrophysical Codes for Exascale

High Performance Computing (HPC) based simulations are crucial in Astrophysics and Cosmology (A&C), helping scientists investigate and understand complex astrophysical phenomena. Taking advantage of exascale computing capabilities is essential for these efforts. However, the unprecedented architectural complexity of exascale systems impacts legacy codes. The SPACE Centre of Excellence (CoE) aims to re-engineer key astrophysical codes to tackle new computational challenges by adopting innovative programming paradigms and software (SW) solutions. SPACE brings together scientists, code developers, HPC experts, hardware (HW) manufacturers, and SW developers. This collaboration enhances exascale A&C applications, promoting the use of exascale and post-exascale computing capabilities. Additionally, SPACE addresses high-performance data analysis for the massive data outputs from exascale simulations and modern observations, using machine learning (ML) and visualisation tools. The project facilitates application deployment across platforms by focusing on code repositories and data sharing, integrating European astrophysical communities around exascale computing with standardised SW and data protocols.

astro-ph.IM↗

arXiv2025

Towards Exascale Computing for Astrophysical Simulation Leveraging the Leonardo EuroHPC System

Developing and redesigning astrophysical, cosmological, and space plasma numerical codes for existing and next-generation accelerators is critical for enabling large-scale simulations. To address these challenges, the SPACE Center of Excellence (SPACE-CoE) fosters collaboration between scientists, code developers, and high-performance computing experts to optimize applications for the exascale era. This paper presents our strategy and initial results on the Leonardo system at CINECA for three flagship codes, namely gPLUTO, OpenGadget3 and iPIC3D, using profiling tools to analyze performance on single and multiple nodes. Preliminary tests show all three codes scale efficiently, reaching 80% scalability up to 1,024 GPUs.

cs.DC↗

arXiv2023

Measurement of the evolving galaxy luminosity and mass function using clustering-based redshift inference

We develop a framework for using clustering-based redshift inference (cluster-$z$) to measure the evolving galaxy luminosity function (GLF) and galaxy stellar mass function (GSMF) using WISE W1 ($3.4μm$) mid-infrared photometry and positions. We use multiple reference sets from the Galaxy And Mass Assembly (GAMA) survey, Sloan Digital Sky Survey (SDSS) and Baryon Oscillation Spectroscopic Survey (BOSS). Combining the resulting cluster-$z$s allows us to enlarge the study area, and by accounting for the specific properties of each reference set, making best use of each reference set to produce the best overall result. Thus we are able to measure the GLF and GSMF over $\sim 7500\, \mathrm{deg}^2 $ of the Northern Galactic Cap (NGC) up to $z<0.6$. Our method can easily be adapted for new studies with fainter magnitudes, which pose difficulties for the derivation of photo-$z$s. The measurement of the GSMF is currently limited by the models for k-corrections and mass-to-light ratios, rather than more complicated effects tied to the evolution of the differential galaxy bias. With better statistics in future surveys this technique is a strong candidate for studies with new emerging data from, e.g. the Vera C. Rubin Observatory, the Euclid mission or the Nancy Grace Roman Space Telescope.

astro-ph.GA↗

arXiv2021

Galaxy And Mass Assembly (GAMA): $\mathbf{z \sim 0}$ Galaxy Luminosity Function down to $\mathbf{L \sim 10^{6}~L_\odot}$ via Clustering Based Redshift Inference

In this study we present a new experimental design using clustering-based redshift inference to measure the evolving galaxy luminosity function (GLF) spanning 5.5 decades from $L \sim 10^{11.5}$ to $ 10^6 ~ \mathrm{L}_\odot$. We use data from the Galaxy And Mass Assembly (GAMA) survey and the Kilo-Degree Survey (KiDS). We derive redshift distributions in bins of apparent magnitude to the limits of the GAMA-KiDS photometric catalogue: $m_r \lesssim 23$; more than a decade in luminosity beyond the limits of the GAMA spectroscopic redshift sample via clustering-based redshift inference. This technique uses spatial cross-correlation statistics for a reference set with known redshifts (in our case, the main GAMA sample) to derive the redshift distribution for the target ensemble. For the calibration of the redshift distribution we use a simple parametrisation with an adaptive normalisation factor over the interval $0.005 < z < 0.48$ to derive the clustering redshift results. We find that the GLF has a relatively constant power-law slope $α\approx -1.2$ for $-17 \lesssim M_r \lesssim -13$, and then appears to steepen sharply for $-13 \lesssim M_r \lesssim -10$. This upturn appears to be where Globular Clusters (GCs) take over to dominate the source counts as a function of luminosity. Thus we have mapped the GLF across the full range of the $z \sim 0$ field galaxy population from the most luminous galaxies down to the GC scale.

astro-ph.GA↗

arXiv2019

The Outer Stellar Halos of Galaxies: how Radial Merger Mass Deposition, Shells and Streams depend on Infall-Orbit Configurations

Galaxy mergers are a fundamental part of galaxy evolution. To study the resulting mass distributions of different kinds of galaxy mergers, we present a simulation suite of 36 high-resolution isolated merger simulations, exploring a wide range of parameter space in terms of mass ratios (mu = 1:5, 1:10, 1:50, 1:100) and orbital parameters. We find that mini mergers deposit a higher fraction of their mass in the outer halo compared to minor mergers, while their contribution to the central mass distribution is highly dependent on the orbital impact parameter: for larger pericentric distances we find that the centre of the host galaxy is almost not contaminated by merger particles. We also find that the median of the resulting radial mass distribution for mini mergers differs significantly from the predictions of simple theoretical tidal-force models. Furthermore, we find that mini mergers can increase the size of the host disc significantly without changing the global shape of the galaxy, if the impact occurs in the disc plane, thus providing a possible explanation for extended low-surface brightness disks reported in observations. Finally, we find clear evidence that streams are a strong indication of nearly circular infall of a satellite (with large angular momentum), whereas the appearance of shells clearly points to (nearly) radial satellite infall.

astro-ph.GA↗