Building a modern solution on Hadoop requires stitching together more than 47 different projects. CDAP solves the complexity of this integration requirement by creating an abstraction layer between the infrastructure of Hadoop and the development of your business applications. In doing so, it minimizes the business risk of your projects by reducing your time to market, time to value, and overall TCO.
Three scenarios for enterprises today:
Just starting out with Hadoop:
- You want to avoid making the mistakes that everyone else has made in implementing Hadoop saving time and making data your company’s competitive advantage.
- Fast-track past all of the development that CDAP has in over 4 years encapsulated within the platform.
- Focus your attention and resources on building the app/solution rather than integrating low level infrastructure.
Have built your first or second Hadoop based solution:
- Following your first project, you want to leverage the investment of your team and reuse as much of the code written in the next projects rather than duplicate efforts. CDAP provides a standard way to easily package your code into reusable containers.
- As you add a second and third Hadoop based app or solution, you see the need to build and maintain your own abstraction layer to reuse the work already created. Your abstraction layer begins to grow and requires it’s own budget and organization.
- As more of your enterprise adopts Hadoop technologies, CDAP provides a common platform for your organization and a means for developing and enforcing standards.
Deep into Hadoop:
- You are already running a Haddop or Spark cluster on-premises, but you have a strategic initiative to leverage the cloud for future workloads, or to be able to burst peak workloads into the cloud. So you need any easy way to enable and run a hybrid cloud environment without the cost and complexity of modifying your existing big data workloads.
- You have a need to support multiple distros, processing engines, or data stores for different partners, customers, or internal stakeholders. Each one requires a different developer expert to code and tune your solution. Switching infrastructure components comes at a huge cost in time and resources.
- You are upside down with your Hadoop team. Instead of a 10:1 ratio of app/solution developers to infrastructure developers, you have 5 Hadoop engineers for every 1 solution developer. You want to stay focused on delivering business value and avoid diverting substantial resources away from your company’s core competence by building an internal development platform. Additionally, experienced Hadoop developers are in high demand and attrition is negatively affecting your organization and ability to complete and maintain big data projects.