About
Subscribe

Old Mutual spices up data with Cayenne

Johannesburg, 13 May 1998

Old Mutual has signed with Software Futures for the Cayenne toolset to help model its Individual Life data. Old Mutual selected Cayenne`s Terrain, GroundWorks and Sharework Manager to underpin the process of developing and managing its business models. Old Mutual data architect Dave Muirhead says the insurance giant has a long-standing commitment to data modelling as a core discipline, but tools previously used had not been able to cope with the group`s enterprise modelling needs. "Our philosophy is to have one model of our data at a conceptual level which is mapped to one or more physical implementations" says Muirhead. "We need one integrated model of our data rather than a collection of many discrete models. Only Cayenne offered us this capability." Old Mutual is committed to data modelling because of the requirements of the life insurance industry, says Muirhead. "When we issue a policy, it must last the lifetime of the policyholder - potentially 80 years or more - so the data and the rules around that data must last as long. That means our designs must be well considered and durable." In addition, the basic design and model must transcend technology generations. For example, Old Mutual policy records which were originaly maintained on the earliest tape-based systems have survived several hardware and software migrations and are now intact in the MVS environment. The Individual Life operations have a particularly challenging environment, adds Muirhead. This complexity is a consequence of: * Sheer volumes: Some 10 million people and companies are associated with the policy databases; * The legacy environment: This is being replaced with new applications written in Amdahl`s ObjectStar (formerly Huron); * The requirement from a business perspective that it be transparent and immaterial where applications are running - the front-end must reflect a consistant business model; * The number and variety of products on offer; * The absolute size of the databases, which handle millions of realtime, online transactions a day; and * The process-intensive nature of these transactions, each of which can involve many complex calculations. Individual Life help manage this complexity by separating the presentation layer (how the users view data) from the data management services (where Cayenne fits in), which provide access to the physical databases. Below this are the IDMS, ObjectStar, software packages, and Windows NT databases and application systems. Old Mutual will define its Individual Life business data model in GroundWorks, with the physical database design in Terrain, and use Sharework Manager to assist with version control. Old Mutual has modelled 80% of Indivdual Life data within the context of four primary data families: client, product, agreement and money. However, data modelling is an ongoing commitment, says Muirhead, the business is changing and expanding and there is an ongoing drive to embed more and more knowledge in the application systems. A long-term goal, Muirhead adds, is to model all the business rules that govern our products "It`s the dream of every life insurer to model and implement its rules as data structures. This will help to give the flexibility we need to cope with the rapid change that characterises the insurance industry," says Muirhead. A key factor in the selection of Cayenne, Muirhead emphasises, was the skills Software Futures was able to wrap around the toolset. "The work we have done with Old Mutual emphasises again just how important it is to put in place the right data disciplines up-front, and the benefits that accrue from such an approach," says Michael Bornheim, data management GM at Software Futures.

Share