Diego Herbin STALDER DIAZ
- diego.s.eik@gmail.com
- stalder@iap.fr
- Research Gate
Visiting Phd Candidate 98bis Boulevard Arago 75014 Paris
Education
MSc Applied Mathematics and Computation
National Institute for Space Research (INPE)
BSc in Electronic Engineering
National University of Asunción
Languages
- Spanish and Guarani (Native)
- Portuguese (Professional)
- English (Professional)
- French (Basic Professional)
Research Interests
- - Cosmological Nbody Simulations.
- - Bayesian Analysis Surface Photometric Analysis
- - Galaxies and the Universe - Environmental Effects on Galaxies
- - High-performance computing (GPUs)
Supported by
About me
I'm a Phd Candidate from National Institute for Space Research (INPE). Currently my research it's focused in two main projects: (i) Bayesian Surface Photometry Analysis and (ii) the study of the Environmental effects on galaxies probed with MAGGIE.
I started my PhD in 2013 under the supervision of Reinaldo R. Rosa and Reinaldo R.Carvalho. Thanks to the Brazilian Science Wihtout Borders Scolarship and Gary Mamon (my advisor at the IAP) i'm working at the IAP as a visitor.
Before, i obtained my BA in Electronic Engineering at the National University of Asunción (UNA). My final year project: Optimal boundary control parareal algorithm for cooling electronics circuits, developed under the advice of Christian E Schaerer. In 2013, I obtained my MSc degree in Applied Computing and Mathematics at INPE with the project: A new gravitational N-body GPU simulator for Computacional Cosmology
In my spare time: I love cooking, riding bicycles and traveling with my wife Mariam :)
Research
MAGGIE is a prior- and halo-based, probabilistic, abundance matching (AM) grouping algorithm for doubly complete subsamples (in distance and luminosity) of flux-limited samples. Previously it was tested on groups extracted from a mock Sloan Digital Sky Survey Legacy redshift survey [1,2]. So this project aims to improve MAGGIE and then test our results on the Sloan Digital Sky Survey Legacy survey.
We are developing a new tool called PyPiGALPHAT (Python Pipelining GALPHAT) to access and analyze efficiently, samples of galaxies (with thousand objects). We analyze process synthetic images and a samples of high stellar mass early-type galaxies using GALPHAT (GALaxy PHotometric ATtributes), a recently developed software for modelling galaxy images based on Bayesian formalism. However GALPHAT takes about fifteen minutes to process only one galaxy and requires a careful definition of the configurations parameters. This poses the challenges of developing and using new tools for data access, and analysis. PyPiGalphat organize the tasks in several steps: the preprocessing, processing and post processing. This new tool allows us the feed a CPU cluster(200 cores) to process and analyze bigger samples. We characterize the bias and the Bayes factor reliability. Related to the Fapesp Thematic Reseach Project
We developed a new N-body simulation environment, named COLATUS_ENVIU, which is a brazilian colaboration between LAC-INPE and IC-UFF, for applications in computational cosmology. The study involves the application of GPU/CUDA technology for simulation of gravitational large-scale structure formation (8 Mpc/h < L < 128Mpc/h), incorporating new features that allow one to test alternative theoretical proposals to the standard cosmological model LCDM. Simulations of the Hubble volume comprising a number of elements in the order of 10^6 are set to the same ranges of redshift, boundary conditions and initial conditions considered in the simulations that do not use the GPU technology. Comparative studies involving the two-point correlation function and the velocity dispersion, calculated also on results obtained from the Virgo consortium, prove the performance of the simulator COLATUS_ENVIU for applications in computational cosmology.