The line between what constitutes a software-as-a-service (SaaS) application and a platform-as-a-service (PaaS) environment has always been relatively thin. SaaS applications that expose an API to third-party developers can quickly transform into a development platform.
The latest SaaS application provider to make that transition is GoodData, which today announced the GoodData Open Analytics Platform. According to Jeff Morris, vice president of marketing for GoodData, the GoodData Platform -- designed as a PaaS environment optimized for data discovery governance -- provides a framework for consolidating structured and unstructured data that can be exposed via GoodData APIs.
Developers have the option of using those APIs to invoke that data or to build analytic applications, written in the R programming language, that will reside directly on the GoodData Open Analytics Platform, says Morris. The data residing on the GoodData platform can be stored as raw Hadoop files or in a data warehouse based on a columnar database.
Morris explains that in addition to pulling data from on-premise systems, the GoodData Open Analytics Platform comes with connectors and metadata maps for 50 sources of data in a GoodData private cloud that spans hundreds of servers, providing access to terabytes of RAM to support analytic applications running in memory.
The Platform, adds Morris, is designed to be extensible in that organizations can make use of ETL tools or APIs to load data into the platform, define and execute their own workflows, and share data with other data discovery tools.
GoodData is not the only vendor making analytics available as a cloud service. But what will distinguish GoodData most is how open the analytics environment is, says Morris. In a world in which organizations can never be sure where their next source of Big Data might be coming from, the GoodData Open Analytics Platform provides a level of interoperability that will allow organizations to more easily correlate multiple data sources.
At the end of the day, Big Data is as much about collecting massive amounts of information as it is about finding the most cost-effective and practical approach to actually analyzing that information. For these reasons, organizations looking to make data-driven decisions based on multiple sources of information are going to look to the cloud not only to store that data but also to run analytics applications as close to that data as possible. Naturally, once that occurs, the next step will be to expose the results of those analytics via APIs to inform other applications -- in effect for the first bringing the benefits of advanced analytics to an emerging API economy that today is all too often starved for truly meaningful data.