Computing – University of Copenhagen

Forward this page to a friend Resize Print Bookmark and Share

Dark Cosmology Centre > Research > Tools >> > Computing

Observations of galaxy or cluster properties must be compared with theoretical predictions. Since the physical processes behind the formation and evolution of such cosmological structures are very complex, we use large numerical computer simulations to make predictions.

At DARK we make numerical predictions for a wide range of observations, including:

  1. Predicting the flux in lyman-alpha radiation for distant galaxies
     
  2. The transfer of lyman-alpha emission through the Universe
     
  3. Predicting the gas and X-ray properties of galaxies
     
  4. Predicting the Sunyaev-Zeldovich effect of galaxy clusters
     
  5. Demonstrating that the gas temperature is determined by the dark matter temperature
     
  6. Establishing the velocity distribution of dark matter particles

Many simulations are performed on the  "Steno" cluster consisting of over 200 nodes of 4 or 8 GB onboard memory. Each node is a dual processor dual core AMD Opteron machine, and has access to a 26 TB disk array over an Infiniband network allowing a node to write at 110 MB/s.

All local servers and laptops are Macs, which run the not so computer intensive tasks.