Last update:
Apr 1, 2000
|
|
|
|
Espresso – a Feasibility Study of a Scalable, Performant ODBMS
|
Dirk Duellmann
CERN
Speaker:
Dirk Duellmann
At the request of the LHC Computing Board, the RD45 project has carried out an analysis
of the risks associated with the proposed strategy for providing object persistency for the LHC
experiments (namely an ODBMS such as Objectivity/DB and a MSS such as HPSS). Although these systems
are being successfully used by a number of experiments, including BaBar at SLAC, one of the primary
conclusions of the risk analysis is that the manpower required to develop a suitable system
in-house should be estimated. This is due to the relatively small size of the ODBMS market –
estimated to be worth some $100M per year – and because of a lack of a convincing commercial
alternative to Objectivity/DB. We describe the goals and non-goals of the feasibility study, the
main architectural features, scalability and performance characteristics and status of the
prototype. We present an estimate of the manpower that would be required to design and build a
full-production system and compare this with alternatives. Finally, we discuss the possibility of
using technologies such as object-relational databases, which some people now argue will become the
dominant database technology in the medium term.
Presentation:  | Short Paper:  |
|
Bulletins
| Home | Bulletins | Committees | Scientific Program | Docs by topics | Social Event | Conference Location | Secretariat |
Privacy Policy |