Recently, we wrote about five reasons to use a NoSQL database to handle today’s volume, velocity and variety of data. But how do you make the move from a relational database to a NoSQL database? Who do you involve and how do you get buy-in?
DragonSpears suggests following these six steps:
1. Determine Your Use Case
Do you have an application need? Do you have data that you can’t use right now, that you’re struggling to repurpose? Are you extracting data, but have to transform it from one silo to another? Make sure to pinpoint your priorities, which can also include data model flexibility, performance or horizontal scalability.
2. Start Small and Build
Put together a proof of concept by creating a very simple application. Use it as the agile environment to start building out. As you ingest the data, DragonSpears can enrich it by using ontologies and public information. We’ll help you ingest it, examine it and create new value by making relationships. For example, healthcare businesses will want to attach public ontologies to give different medical terms more value, instead of treating them as text.
3. Choose Your Champion
Switching from a relational database to NoSQL involves a culture change at different levels of management. The ideal champion will have influence across your organization and the ability to convince developers as well as upper decision-makers in your company.
4. Engage and Train Your Team
Many developers have a long history with relational databases. Get them excited to work with NoSQL. Encourage them to see it as an opportunity, not a threat. Invite them to voice their concerns. Make sure to get consensus on specific NoSQL databases so there is collective ownership.
Most NoSQL databases are open source and have communities that help new users get started. For those who want more than hands-on training, instruction is available through online forums, classroom-based learning and free online information.
5. Hire a Full Stack Developer
Make sure this individual can handle front end, business logic and back-end programming. This person doesn’t have to be a super database administrator to be effective—he or she can store the information without robust knowledge of how the database works.
6. Plan Carefully
NoSQL databases are so easy to work with that developers tend to dive in headfirst, without taking the time to create a sound data model, which can cause problems down the road.
Ask DragonSpears to show you a NoSQL proof of concept—we can deliver one in as little as one week. We’ll help you maximize the value of the data you collect and help you move at the speed ofbusiness.
