Tootfinder

Optimizing Cloud-to-GPU Throughput for Deep Learning With Earth Observation Data
Akram Zaytar, Caleb Robinson, Girmaw Abebe Tadesse, Tammy Glazer, Gilles Hacheme, Anthony Ortiz, Rahul M Dodhia, Juan M Lavista Ferres
https://arxiv.org/abs/2506.06235

Optimizing Cloud-to-GPU Throughput for Deep Learning With Earth Observation Data
Training deep learning models on petabyte-scale Earth observation (EO) data requires separating compute resources from data storage. However, standard PyTorch data loaders cannot keep modern GPUs utilized when streaming GeoTIFF files directly from cloud storage. In this work, we benchmark GeoTIFF loading throughput from both cloud object storage and local SSD, systematically testing different loader configurations and data parameters. We focus on tile-aligned reads and worker thread pools, usin…

AT2019cmw: A highly luminous, cooling featureless TDE candidate from the disruption of a high mass star in an early-type galaxy
Jacob Wise, Daniel Perley, Nikhil Sarin, Tatsuya Matsumoto, K-Ryan Hinds, Yuhan Yao, Jesper Sollerman, Steve Schulze, Aleksandra Bochenek, Michael W. Coughlin, Kishalay De, Richard Dekany, Sara Frederick, Christoffer Fremling, Suvi Gezari, Matthew J. Graham, Anna Y. Q. Ho, Shrinivas Kulkarni, Russ R. Laher, Conor Omand, Natalya Pletskova, Yashvi Sharma, Kirsty…

AT2019cmw: A highly luminous, cooling featureless TDE candidate from the disruption of a high mass star in an early-type galaxy
We present optical/UV photometric and spectroscopic observations, as well as X-ray and radio follow-up, of the extraordinary event AT2019cmw. With a peak bolometric luminosity of ~$\mathrm{10^{45.6}\,erg\,s^{-1}}$, it is one of the most luminous thermal transients ever discovered. Extensive spectroscopic follow-up post-peak showed only a featureless continuum throughout its evolution. This, combined with its nuclear location, blue colour at peak and lack of prior evidence of an AGN in its host …