Big Data

Ontology 4 handles big data variety and provides insight by linking large data sets to existing enterprise data.

Big data presents technical challenges for organisations as they try to derive value by gaining insight into massive, varied and fast changing data sets. The problem doesn't just stem from the volume and velocity of the data sets themselves, but also from the variety challenge posed by gaining big data insight in the context of the rest of organisation's enterprise data. Analysing a large data set to find a specific business entity such as a customer may be of little value unless the enterprise's other customer data sources can be linked and included in the analysis. Ontology 4 is specifically designed to link data from different sources and provide a choice of access methods including "on demand" data fetches and mixed-source searches. This means that big data sets can be fully exploited in the context of their relationships with other enterprise data.

Linking data for analysis like this used to be seen as an important function of data warehousing solutions, but the scale of big data sets, the multitude of different incoming data formats and the rate of change in the enterprise infrastructure are stretching the technical capabilities of traditional technologies. The greatest challenge that organisations face in embracing the exciting business transformation opportunities that big data presents, is not necessarily dealing with the volume of incoming data, instead it comes from the cost and time required to accommodate the variety of new sources and formats of data. It is the variety "v" that is hardest to address.

As organisations explore different ways to engage with their markets and stakeholders, so the number of potential data sources and data formats increases. Before any value can be derived from social media interactions, cloud-based systems or even server and application log files, a lengthy and costly data integration project will be required to access and link that data. Ultimately, these logistical and budgetary restraints will constrain the value of big data to the enterprise. To fully exploit big data sources, adding new and varied data sources must become a low-cost, routine process.

Ontology 4's graph-based, schema-less approach utilises an inherently agile semantic model to link data from different sources. Adding additional data sources and linking the entities described in those data sources to existing enterprise data is achieved by simply extending the data linkages described in the semantic model. Unlike alternative approaches that require significant upfront analysis and create a rigid and difficult to modify schema-based model, Ontology 4's semantic model simply evolves to accommodate new data linkages. This ability means that new data sources can be connected and exploited in a fraction of the time and cost of traditional methods.

As sources of data are linked in this way, so a logical data warehouse is created, project by project. The logical data warehouse doesn't necessarily physically store the data, but instead can link and search different data sources to deliver insight. Ontology sees the logical data warehouse as the technical solution to linking next-generation, big data management platforms with existing structured and unstructured enterprise data. Ontology 4's graph-based semantic search platform provides the foundation for the logical data warehouse – a new way of seeing enterprise data.

Read more about Ontology's approach to big data in the Ontology Big Data and the Logical Data Warehouse Whitepaper here.