AI and Analytics in Production

Chapter 6. Tips and Tricks

So, what does it take to put artificial intelligence (AI), machine learning, or large-scale analytics into production? In part, it depends on decisions you make as you design and implement your workflows and how you set up your cluster(s) to begin with. The technologies available for building these systems are powerful and have huge potential, but we are still discovering ways that we can use them. Whether you are experienced or a newcomer to these technologies, there are key decisions and strategies that can help ensure your success. This chapter offers suggestions that can help you make choices about how to proceed.

The following list is not a comprehensive “how-to” guide, nor is it detailed documentation about large-scale analytical tools. Instead, it’s an eclectic mix. We provide technical and strategic tips—some major and some relatively minor or specialized—that are based on what has helped other users we have known to succeed. Some of these tips will be helpful before you begin, whereas others are intended for more seasoned users, to guide choices as you work in development and production settings.

Tip #1: Pick One Thing to Do First

If you work with large volumes of data and need scalability and flexibility, you can use machine learning and advanced analytics in a wide variety of ways to reduce costs, increase revenues, advance your research, and keep you competitive. But adopting these technologies is a big change from conventional computing, ...