One advantage of using MongoDB for data science is that it is trivial to get data into MongoDB, and easy to change your data model as needed to support analysis. With our data in MongoDB, we can take advantage of indexes, the MongoDB query language, and the aggregation framework to manipulate and explore data efficiently, as we work to better understand and refine our requirements. MongoDB provides a free tier option in Atlas, which is perfect for getting started. Atlas is a MongoDB hosting service provided by the people who built MongoDB. It's easy to migrate data from the free tier to production scale databases when you're ready. To get started, please sign up for MongoDB Atlas, visiting the link provided in the lecture notes for this lesson. We'll be loading some initial data sets that we will clean up and begin analyzing as soon as we're done creating an Atlas cluster. Please create an account and save your credentials somewhere. Once you've submitted an account creation form, you'll be prompted to create a group. Atlas uses groups to enable you to control who has access to your data. As is often the case, you can grant full read write access to some users such as the data scientists on your team and read only access to others. Please enter analytics as the group name. Throughout this course, we'll give you pretty specific instructions on set up tasks such as this so that if you have trouble, it's easy to diagnose the issue. Once you've created a group, you'll be taken to a screen like this. Please note that as with all websites, Atlas is actively developed. The user interface you see may look a little different from what we see here. Note also, that Atlas uses Intercom, and you might see a welcome message when you first create an account. Now, don't be nervous about the fact that you see a cost here under pricing. Just select M0 here, under instance size, and you'll see the price go to zero. Also, let's name this cluster mflix. This is the name of the app we'll be building as we move through this course. Once you've named the cluster and selected M0 as the instance size, please scroll to the bottom of this modal. And create an administrative user with the username analytics, and the password, analytics-password. We'll use these credentials to connect to this MongoDB cluster from Python scripts and applications later in this course. Now, if the term cluster is unfamiliar to you, don't worry. All you need to know at this point is that a cluster provides redundancy. The cluster we are creating is actually three MongoDB database servers. This a safeguard to make sure your data is always available to you. While the probability is small on something going wrong on any given day, computers crash, network connections break, and other problems emerge. Having a cluster rather than just a stand alone instance ensures that you'll be able to access your data even if something bad happens to one of the servers in our cluster. The client libraries we will use from Python will automatically switch to one of the other servers if for some reason it could no longer communicate with the one it was reading from and writing to. This is called automatic failover. All nodes in the cluster maintain copies of your data, so it really doesn't matter which one your application uses. All this is handled for you automatically. You probably won't even notice of a failover happens. Okay, with that little bit of context about what a cluster is, click Confirm & Deploy to initiate the creation process. You might see a captcha pop up. Simply do what it asks you to do, verify. It will take a minute or two for your clusters to be created. While we wait, let's ensure we can access our cluster once we begin to write a little Python code. Click the Security tab, then click the IP-Whitelist tab. Click the Add IP Address button And then the Allow Access from Anywhere button. And finally, Confirm. Atlas allows us to restrict access to a cluster to specific IP address ranges to, for example, ensure only computers in a specific corporate network can connect to the cluster. Allowing access from anywhere will make it easy for you to access MongoDB from wherever you are and simplifies access for this course. You will still need to supply a username and password to access the cluster. Now, click the Clusters tab and wait for your cluster to finish initializing.
