Thursday, October 15, 2020, 11:00am to 12:00pm
Jun Makino (Kobe University) "Title: Frameworks for large-scale astrophysical simulations"
In the last three decades, it has been becoming more and more difficult to develop applications which can achieve high, or even modest, efficiency on advanced HPC systems. The development of highly efficient and highly scalable software for modern HPC systems like K or Fugaku requires multi-year effort of a team of researchers with exceptional talents. We have been working on a new approach to solve this problem through the use of framework. Our FDPS (Framework for Developing Particle Simulators) framework allows the users to write their programs in the way very close to the way they would write them for a single-processor machine, and run them on large-scale parallel systems with very high efficiency. The guiding principle for the
design of FDPS is to minimize inter-node communication and main memory access per timestep, without increasing the floating-point operations
too much. I will discuss current state of FDPS and applications based on FDPS. If time allows, I'll discuss how we can make the design of computer architecture "scientific" or quantitative in the true sense.
Wenbin Lu (Caltech) "A Unified Picture of Fast Radio Bursts"
Abstract: Fast radio bursts (FRBs) are short duration (~ms), very bright, radio transients. Their detection a decade ago was a major unexpected discovery in astronomy in decades. Hunting for FRBs and measuring their physical properties have become one of the leading scientific goals in astronomy. This effort has led to a rapidly growing sample with extremely diverse properties in luminosity (10^38 to 10^45 erg/s), duration (0.1 ms to 10 ms), and repetition rate (some objects have multiple bursts in an hour and many just one burst in a few years). I will present a study of their cosmological volumetric rate density and provide evidence that these bursts all belong to the same class of transients --- most likely all are repeaters. According to my model, disturbances close to the surface of a magnetar launch Alfven waves into the magnetosphere, which propagate to a distance of a few tens of neutron star radii and then produce coherent radio emission. The coincident hard X-rays associated with the Galactic FRB 200428 can be understood in this scenario. This model provides a unified picture for weak Galactic FRBs as well as the bright bursts seen at cosmological distances. If time allows, the polarization properties of FRBs will also be addressed.