Object-oriented databases the next wave in complex data management
“Manufacturers want to connect their factory-floor operations and their production systems up to the top of the enterprise, into their upper business systems,” says Dick Slansky, senior analyst, product life-cycle management (PLM) & discrete industries for Dedham, Mass.-based ARC Advisory Group .
“Manufacturers want to connect their factory-floor operations and their production systems up to the top of the enterprise, into their upper business systems,” says Dick Slansky, senior analyst, product life-cycle management (PLM) & discrete industries for Dedham, Mass.-based ARC Advisory Group . “The challenge is capturing that data in its different formats.”
Relational databases offer one approach to complex data management, but they are single server-oriented and constrained by tables. While their strength lies in ad hoc queries, relational databases don't remember past queries. “It's always an atomic operation,” says Todd Stavish, SE manufacturing for Objectivity, a provider of high-performance database engines.
Stavish says Objectivity offers another approach: object-oriented data management platforms. These platforms, such as the Objectivity/DB solution, collect data—e.g., flat files, spreadsheets, numbers, and raw data—from information silos and disparate systems, and then store the data as objects in a repository, making the data appear as if on a single server. This sort of structure enables vertical and horizontal interoperability so that data can be correlated for predictive analysis.
“In this federated architecture, data stays at its source, but horizontal levels and higher levels can actually see that data,” says Brian Clark, VP, technical services, Objectivity. “It's what we call the single logical view across the federation.”
Unlike relational databases, “You aren't forced to make the world look like a table,” says Stavish. Instead of querying a whole database, developers can open one database and structure it any way they choose. Using data fusion, smart relationships are created so that the same queries don't have to be run again and again.
Manufacturers are recognizing the benefits of object-oriented databases.
Erlangen, Germany-based Siemens Power Generation uses Objectivity/DB as the storage and management knowledge-base repository for its compressor engineering system (CES) application, which matches compressor performance and rotor dynamics to customer-specified applications. The CES process includes complete technical selection and design calculation—something achieved through the Objectivity/DB platform since it connects data housed in multiple disparate systems. More than 300 Siemens workers are registered to use the system, with an average of 40 to 50 users working concurrently around the world in CES.
In manufacturing's flat world, says Slansky, “Object-oriented databases enable knowledge capture and reuse, which will be come extremely important in the next five to 10 years.”