Learn what the big data ecosystem looks like and how difficult it may be to implement such a solution.
- [Ben] Let's talk about the myth that big data will instantly yield great insights for your business. And I want to introduce this as a way of understanding the actual timeline of what a big data implementation looks like. In the first stage, you're just planning things out. You're trying to understand the needs that led you to want to go down this path, and then figuring out which platforms will help you with those needs. After you've surveyed the land, you'll want to do an evaluation to see if indeed the platforms on your short list will actually help you meet those challenges you face.
After you've chosen a platform, it's time to figure out the architecture you'll need, and begin the procurement process. Sometimes, if you're choosing things in the cloud, that can help with this step, because once you decide that you want something, you click a couple buttons, and boom, your hardware is up and running. But there's still a lot of work to be done. With all of that complete, it's time to implement the actual architecture. This may require installing physical hardware, connecting it to your network, installing the platform software, such as Hadoop, and then configuring everything to talk to each other.
Don't forget about security either. Security is a major concern, and really should be at the forefront of your implementation phase, even going back to the planning and setup phases. Lastly, you'll finally be able to use your big data platform, and with this, you need good developers that understand the platform, and folks that have figured out which projects would be the biggest return on investment for you to work on first. So finally, after typically months of the previous steps, you're finally realizing value. So the myth that big data, we'll just sprinkle it onto our existing infrastructure, and bam, we've got insights, really isn't true.
This is a journey, and it's going to be a long journey. And as we talk more about big data, you'll understand the benefits, and also some of the limitations.