Computationally challenging experiments such as the STAR at RHIC (Relativistic Heavy Ion Collider [1]) located at the Brookhaven National Laboratory (USA), have developed a distributed computing approach (Grid) to face their massive computational and storage requirements. Since the peta-bytes of data produced by STAR are geographically spread, it is necessary to face the question of efficient data transfers and placements in order to bring requested dataset to a particular site for further analyses. Our aim is to create a plan how to transfer data from data-warehouses to the requested site in the shortest time. The presented approach is based on Constraint Programming with initial modeling idea from [2] further extended in [3]. This paper brings a new search heuristic used during the selection of routes.
[1]
H. M. Spinka,et al.
Experimental and Theoretical Challenges in the Search for the Quark Gluon Plasma: The STAR Collaboration’s Critical Assessment of the Evidence from RHIC Collisions
,
2005
.
[2]
Helmut Simonis,et al.
Constraint Applications in Networks
,
2006,
Handbook of Constraint Programming.
[3]
Ion Stoica,et al.
Peer-to-peer systems II : second International Workshop, IPTPS 2003, Berkeley, CA, USA, February 21-22, 2003 : revised papers
,
2003
.
[4]
Roman Barták,et al.
Using Constraint Programming to Plan Efficient Data Movement on the Grid
,
2009,
2009 21st IEEE International Conference on Tools with Artificial Intelligence.