ANR-B3DCMB is an interdisciplinary project funded by the French National Research Agency (ANR) and aiming at addressing some major challenges in data analysis of current and forthcoming data sets of cosmic microwave background (CMB) anisotropy data. Modern CMB data analysis requires cutting edge computational techniques and numerical algorithms as well as efficient statistical approaches in order to solve the challenges it faces already or will face soon. All these aspects have to be addressed concurrently, informing each other, while being guided by insights into the physics of the problem and building on the hands-on experience from the previous data analysis efforts. This is this holistic approach, simultaneously accounting for points of view of all these key science areas, which is the cornerstone of the B3DCMB project.
Consequently, the B3DCMB project will infuse data analysis of CMB data sets with novel techniques motivated by the cutting edge research in high performance scientific computing and appropriate for the most promising statistical approaches, which we will select not only on the basis of their statistical merit but also considering whether they permit efficient computational solutions and implementations appropriate for novel architectures of current and anticipated the largest massively parallel supercomputers.
While aiming for this goal, we expect that the CMB data analysis problems will motivate new ideas or even lines of research in high performance scientific computing and statistics.
Later, with those new tools at hand we plan to contribute to our understanding of modern cosmology via actual analysis of cosmic microwave background data sets. The experiments of our particular interests are: ground-based efforts, Simons Array, Simons Observatory, and a satellite mission, LiteBIRD.
Continue to the next pages (or use the manu above) to find out more about the project, its science, and the team we have set up to implement this program.