Last update:
Apr 1, 2000
|
|
|
|
|
| on same keytopic |
  |
|
|
|
An Active Model Approach to Data Analysis in the Petabyte Era
|
G. Chevenier1,
C. Koch1,
J.-M. Le Goff1,
I. Willers1,
R. McClatchey2,
Z. Kovacs2,
P. Brooks2,
M. Zsenei3
- CERN, Geneva, Switzerland
- Centre for Complex Cooperative Systems, Univ. West of England, Frenchay, BS16 1QY UK
- KFKI, RMKI, 1525 Budapest, Hungary
Presented by:
Jean-Marie Le Goff
Experimenters at LHC must equip themselves with the
appropriate software tools and techniques to store and
access petabytes of data in a manner that facilitates
physics analysis in a timely and efficient manner. The issue
of storing petabytes of data has largely been resolved by
the emergence of data base products, both relational and
object-oriented in nature that can handle such data volumes.
Accessing petabyte data warehouses to allow concurrent
analysis by physicists in multiple Regional Centres
distributed across continents with adequate response is
quite a different matter and one which is, as yet,
unresolved for LHC experiments. This paper investigates this
issue and adopts a truly object-oriented approach in
proposing an active object model as the basis for
distributed large-scale data analysis. This integrated model
-based approach is contrasted with the traditional approach
of replicating data between multiple, independent data
analysis Centres and conclusions are drawn as to the
advantages of using an active data model approach.
|
Bulletins
| Home | Bulletins | Committees | Scientific Program | Docs by topics | Social Event | Conference Location | Secretariat |
Privacy Policy |