The Impact of Data Fabric Architecture on Single Source of Truth

Once upon a time, the implementation of single source of truth was typically to build an all-encompassing data warehouse. ??That has the obvious advantage of having a single source of truth from a single central location.? However, it has the disadvantage that the store becomes a behemoth data warehouse, where data must be synchronised and replicated from its true source.? This is complex to orchestrate and manage and provides potential issues for timing and distortion.

Data Fabric architectures can provide a more elegant solution which solves these disadvantages.? The implementation changes such that data sets across the domain may have different physical sources of truth. ?For example, front office data such as trading models, orders and execution data are sourced from the front office system.? That is the real source of truth.? Performance data will be from the performance system and tax holdings information from the Fund Accounting source and so on.?

The Single Source of Truth can become the Data Fabric technology itself that allows data sets to be published from a virtualised source that, under the hood, combines data across not only different physical sources of truth, but across different systems and technologies.? Whether it be SQL Server on Azure, Oracle, Snowflake, or via an API or even a flat file etc.? It is this interoperability where data sources can be plumbed into the fabric, without code, that differentiates genuine Enterprise Data Management solutions from other more antiquated architectures that simply combine Data Warehouses with hard coded integration.?

Curium Data Systems has developed Data Fabric Plus as part of our CuriumEDM solution.? It is an extension of our philosophy that data should not be moved unless necessary and one which our Data Quality solution, CuriumDQM, has been operating under for over a decade.? This allows our clients to utilise data in-situ, without replication. The list of data technologies supported has grown as technologies have emerged.? Utilising Curium’s well proven capabilities to combine data across multiple sources and technologies and provide value added data sets, CuriumEDM has been extended to allow these virtualised data sets to be consumed by third party tools such as Power BI, Tableau or many other technologies via an industry standard interface. ?All the hard work is done once within CuriumEDM to model and map physical data sources, without code, so the BI developer and data scientists can get on with solving their visualisation or analytics problem and not waste time data wrangling. ?CuriumEDM can operate as a virtualised single source of truth.??

Importantly the “data fabric” capability is combined with Curium’s ability to provide data quality, master data management, integration and data governance, hence the “Plus” in the name. If you can combine great control with great capability, then it is a compelling solution.

If you think this sounds interesting, please reach out to us.?