resume research album links contact
Education
  • Ph.D(WashU)
  • B.S(USTC)
  • Research
  • Numerical Relativity
  • High Energy Astrophysics
  • HPC
  • T.A.
  • PHYS 117
  • PHYS 118
  • PHYS 206
  • PHYS 217
  • PHYS 312
  • PHYS 427
  • PHYS 505
  • PHYS 529
  • PHYS 546
  • Skills & Fun
  • Clustering
  • Photography
  • RESEARCH EXPERIENCES
    nasa.jpg
    Numerical Relativity (NR)

    Numerical Relativity, like many other fields of computational physics, is trying to solve physics problems numerically. The goal of NR is to study the dynamics of spacetime by solving full Einstein's Equation (EE). Due to the complexity of solving full EE, Newtonian and Post-Newtonian approximations are often used for doing weak field analysis, while NR is necessary to analyze strong gravitational fields around neutron stars and blackholes.
    jet.jpg
    High Energy Astrophysics

    There are a lot of very interesting astrophysical problems arround AGN(Active Galactic Nuclei). People believe there is one or several super massive astrophysical objects at the center of every galaxy. They could be blackholes, neutral stars and even those haven't got names yet. One way to study the intrinsic and dynamic properties of those objects is to detect the high energy radiations(X rays, Gamma rays) from those objects. Another way is to do theoretical analysis based on existing theories and experiments. Numerical simulations work as bridges between them.
    What I am doing is to simulate two colliding jets and study the radiations emited. My work is based on two existing codes. One is able to do SRMHD(Special Relativistic Magnetohydradynamics) simulations of relativistic jets and the other one is able to calculate radiations from known physical data. By considering many micro physics phenomena which are supposed to be triggered in that condition, we will be able to tell which micro physics phenomenon is more important to generate observed data and in some sense determine the basic mechanism causing certain astrophysical phenomenon.

    o2k.jpg
    High Performance Computing (HPC)

    HPC is an imperative part of numerical relativity, because it is almost impossible to solve Einstein's Equations with personal computers at least now. I started learning HPC by setting up a 8-node cluster(hydra) and running codes on it. At the same time, I also played with the SGI origin 2000 supercomputer(64 x R12000 MIPS Processors) at WashU Center For Scientific Parallel Computing. To have more fun, I participated The Global Grid Testbed Collaboration. First, I installed grid-in-a-box downloaded from NCSA and configured it to work. Then I applied to get an account on NCSA machines and a NCSA Certificate. That was my first time playing with grids computing. That experience expanded my visions in HPC. Tomorrow's HPC is not only writing parallel codes and running them on some supercomputers. It will be more collaborative and general. Everybody will be doing HPC in the near future without noticing it.
    << Prev:Teaching Next: Skills & Fun >>

    email Save money and money will save you
    Monday November 23rd 2009 09:04:57 PM
    Copyright © 2003-2006 Jian Tao