Data architecture, similar to Application or Infrastructure architecture has been evolving rapidly due to innovation and speed of adoption. From databases which handle monolithic applications, the data platform industry fast tracked towards data appliances to host a variety of applications that require huge volume of data. The speed and variety of data that is generated by organizations today, has paved way for new data platforms like Hadoop, NO SQL, In Memory etc.
What is Composite Data Platform Architecture?
A composite or hybrid data platform is something which requires multiple specialized data platforms, which can co-exist for the enterprise data management. There is no single data platform that can fulfill the assortment of business use cases which a data architect faces today. Today, the business users are demanding platforms that can deliver analytic solutions and outcomes with different SLA, veracity, speed etc.
Composite Data Platform Architecture - Why is it required?
From the data management perspective, a composite data warehouse architecture gives ample flexibility to prune data transformation process, expand data handling capability, support new business requirements, achieve faster time to market and moreover freedom from single vendor lock in.
From an investment standpoint - the answer differs for different organizations. An organization which is yet to invest on a scalable data warehouse still has time and money to invest on multiple data platforms that will make their data environment diverse and transformational. However, an organization which doesn’t possess a matured data warehouse, need not worry about investing on multiple data platforms upfront. At the same time, it is always wise for the enterprise architects to consider the business strategy and the road map ahead. That way, introducing new platforms in to the architecture stack would mean only optimizing and not re-engineering.