E & P data resides in multiple, dispersed databases often on different architectures, operating systems and hardware. Data access is further complicated in a multi-vendor environment. Two “solutions” have been advanced to get around these difficulties. You either chose a single corporate and project database and migrate all the relevant data to the selected database. Or you can select a single software vendor who can assure the interoperability of all of the important E & P applications.
In fact neither of these solutions has proven to be satisfactory. Legacy databases are rarely depopulated or decommissioned because the costs and time required for the task are hard to justify. The single-vendor solution has the side effect of eliminating vendor competition leading to unresponsiveness. Single-vendor solutions also tend to restrict the technological innovation that comes from the independent software house.
Paradigm has built common well data server based on CORBA (Common Object Request Broker Architecture) technology that provides a universal solution to these problems and offers access to databases regardless of vendor, physical location, or computer operating system. We recognized that no single database can ever be optimal for storing the diversity of data that are generated and used by the E & P industry. Application software must be able to access multi-vendor, distributed data stores that are supplied by multiple vendors without the need for replication.
Application software must be designed so that it does not have to be modified to access new data types. A database must be extensible to new technology data types.
When Paradigm purchased Geolog from Petroleum Technology Mincom (PTM) it gained access to a very flexible data structure which has led most of Geolog’s clients to adopt it as their permanent repository for well data. In 1996 PTM had assumed responsibility for maintaining and enhancing ConnecT, Chevron’s data bridge between Geolog, OpenWorks and GoCAD. Over 1200 licenses of ConnecT are currently in use. This technology enabled the development of well data servers for a variety of application software and project databases.
CORBA was extended by two powerful facilities. One is the “dynamic data mapping engine” that allows the organization to have complete control of the way that well data is represented. This facility allows for name and value substitution, data selection and units of measure issues. The second addition was the “project name server” which locates project well data regardless of physical location. The well data server lets users access data wherever it resides as long as he or she has the appropriate privileges. Multiple databases can be accessed simultaneously. For example, users can simultaneously acquire tops, horizons, and survey data from OpenWorks, log traces and core data from Geolog, and image log data from Recall. All of the data would appear to be coming from a single virtual data store.
CORBA provides benefits to the IT organization. Clients and servers can operate on any platform or operating system and those platforms can be different from each other. The clients and servers are also language independent. A client or server can be written in a combination of Fortran, C, C++, Java, or virtually any other language. Another intriguing characteristic of CORBA is the ease with which “thin clients” can be written in Java (e.g. a well-log browser for use on a corporate Intranet).
Real Soon Now
Paradigm will continue to offer the ConnecT Data Bridge in addition to a family of CORBA-based well data servers. The data server for OpenWorks 98+ and 2000 is now being commercially licensed. A GeoFrame Data Server is under development and will be released commercially in the third quarter of 2000 as part of the Geolog 6.4 commercial release. GoCAD and Recall data servers are high priority in our development schedule. Paradigm will also release (for testing) its first CORBA based well data server for its seismic applications in June 2000.
© Oil IT Journal - all rights reserved.