Distributed radio astronomical computing
The infant Universe (its first-billion years) remains its least explored era. (HI) emission offer many new insights into the infant Universe. In particular, by observing these at higher redshifts or at low radio frequencies, we are able to get a glimpse of the early history and evolution of the Universe. It is possible to observe these ancient HI intensity fluctuations using extremely precise low frequency radio interferometry, provided that all technical challenges are met. The DIRAC project has developed highly distributed and scalable software to tackle the most demanding aspect of data processing to achieve this detection. This software is already being deployed to process Tera-bytes of data taken by the LOFAR radio telescope and the first upper limits in the signal power levels are already published. The knowledge acquired by DIRAC is also made available to other disciplines like machine learning, for instance to improve the training speed. istributed optimization applications.
Improving the AARTFAAC processing pipeline
Unlocking the LOFAR Long Term Archive
Access and acceleration of the Apertif Legacy exploration of the radio transient sky
Optimized data handling for observations in astronomy
The evolution of embedded star clusters
An eScience infrastructure for huge interferometric datasets
Kernel Tuner greatly simplifies the development of highly-optimized and auto-tuned CUDA, OpenCL, and C code, supporting many advanced use-cases and optimization strategies that speed up the auto-tuning process.
Fast, memory efficient and GPU accelerated radio interferometric calibration program