Project Synergy; PDM Interviews Andrew Lloyd of Oracle Corp. (February 1999)

Andrew Lloyd, Senior Industry Director, of Oracle Corp. describes the intent and scope of project Synergy in an exclusive interview for with PDM.

PDM - We cheekily suggested that Oracle had abandoned E&P when you didn't show up at the New Orleans SEG last year. Had you really left us?

Lloyd - Not at all, we were actually preparing the Offshore Technology Conference. The upstream is a very important business to Oracle - we estimate our share of the upstream database market at around 84 %.

PDM - What were the key drivers behind project Synergy?

Lloyd - With the oil price at a 25 year low and with the proliferation of mergers, both of oils and service companies there is a greater focus on joint ventures, asset swaps and risk sharing in general. This activity generates a legal and practical need to share data; between companies, between departments and between disciplines. This may be to facilitate data sharing during the handover from subsurface to engineering, or again to make legacy data, locked up in filing cabinets 'live again' by distributing it to asset teams and between applications.

PDM - And what exactly is novel about the technology?

The key technology in Synergy is Oracle 8i - the new internet-enabled version of the Oracle Object database. Project Synergy will have at its core, a brand new implementation of industry standard data models from the Petrotechnical Open Software Corp. (POSC) and POSC/CAESAR. These data models will be encapsulated into Oracle Cartridges which are plug-in, domain-specific object-extensions for the Oracle RDBMS. The two new cartridges will be developed by Oracle's Server Technical Development Group.

PDM - so this would be the long-awaited shrink-wrap version of Epicentre?

Lloyd - Yes indeed.

PDM - And the intention is to have a single 'mega' corporate database, or would Synergy offer Oracle Cartridges to each department - and if so, how will you maintain synchronicity.

Lloyd - The underlying Oracle architecture helps us to be able to support either highly centralized or decentralized business and/or computing models. It is very likely that a large enterprise would wish to have reference data in one place and departmental data in another. This is essentially something all Oracle users get when they buy Oracle8; clearly, this will be useful when we deploy Project Synergy products.

PDM - Who exactly is the targeted end-user - the technologists or managers? Geoscientists or finance?

The targeted end-user of Synergy is, ultimately, everyone in the company! Domains to be encompassed include exploration, drilling, reservoir engineering, production and facilities. In one, or possibly two repositories, there will be all the data from seismics and geology, subsurface data, the Shared Earth Model, commercial data, facilities engineering, production.

PDM - What sort of characteristics will Synergy applications offer?

Lloyd - Applications are to include Decision Support Systems (DSS), Web-based data mining, the fusion of technical data with information in Enterprise Resource Planning applications such as those from SAP, PriceWaterhouseCoopers and Oracle Financials.

PDM - The scope of Synergy is a little unnerving, are you going to do the whole thing all at once?

Lloyd - The initial focus will be on two domain-specific areas and should allow for proof of concept before wider deployment. One area will be facilities, through Statoil's Odegard project, another will be drilling optimization and will involve a collaboration with the Mobitech group.

PDM - Epicentre has been criticized in the past for its complexity; and its performance as an operational data store has been questioned. How is Synergy going to address such issues?

Lloyd - Our intention is to provide a practical, efficient implementation of POSC Epicentre which will scale across the largest E&P enterprise whether centralized or not.

PDM - In the 'Doing objects with Oracle 8' (see PDM Vol 3 N 11), Doug Benson explained that Oracle 8's object implementation was a kind of pragmatic subset of OO technology designed to offer a limited OO implementation that would actually work. So is it fair to say that the Epicentre Cartridge will be a compromise between the very inherited and recursive design of the Logical model and the 'relational projection' of current implementations?

Lloyd - Before Oracle8i, our technology along with everyone else's could not implement the model as intended. We will be challenging the model where we don't think it makes sense. We are employing POSC on our development and providing feedback on elements of the standard which we believe could be improved. Our intent is to provide a performant, useful implementation which remains as faithful to the model as makes technical and commercial sense. We hope that our implementation will not only prove the value of the model but also provide momentum to drive its future evolution.

Click here to comment on this article

If your browser does not work with the MailTo button, send mail to pdm@oilit.com with PDM_V_3.3_9902_10 as the subject.

© Oil IT Journal - all rights reserved.