MongoDB Atlas
I have been using MongoDB for a few years now for various personal projects and I really enjoy the features and flexibility it offers me as a developer. I have taken several great classes on, and from, MongoDB in that time frame as well. As with any technology not only is there a lot to learn in general, but there is also a lot to continue to learn to stay up to date and relevant. One of the subjects taught is about MongoDB’s availability and ability to scale both vertically (more RAM, IO capacity, and CPU availability) as well as horizontally (sharding). Another subject that is heavily taught is the idea of clusters and replica sets.
Now since I am just a single developer trying to learn technologies, increase my skills, and develop best practices when using them, I personally do not have the resources to purchase multiple servers, constantly increase their hardware, or have them distributed across the country to understand latency issues. I can setup a replica set on the same laptop on which I write my blogs, to get a basic understanding of what is happening but I think a few things get lost in the learning process when boiling things down to a single Intel processor and no network connections between servers.
What options does one have, either to learn about and use some of these “big boy” features of MongoDB, or to prototype applications using real world settings? MongoDB launched a database-as-a-service (DBaaS) platform called Atlas in 2016 and is an opportunity to utilize many of the “big boy” features without having to personally fund a server farm in my garage. I can’t tell you how excited I was about this. Now I could start clusters in different configurations, set up and use multiple replica sets and perform on-the-fly changes to my test servers with ease. I still had to pay a usage fee though, so I would set up servers, run a few tests, then tear down the servers to avoid any long lasting fees. As such I was able to get practice with the technology, but it posed a challenge to showcase some of my projects connecting to a live database.
Enter March 2017… MongoDB announced an awesome feature and offering within their Atlas product. A free cluster tier called M0. This allows anyone and everyone who has been interested in using Atlas to utilize MongoDB in the cloud, on a clustered environment without even having to give them a credit card. Talk about a great offering for people wanting to use this exciting technology.
Now that we can have access to Atlas and a free cluster, what can we do with Atlas? What’s all the excitement behind it? Well, first off it provides an incredibly intuitive GUI to set up a MongoDB cluster in a variety of regions and with a variety of features. As one might expect the free M0 cluster doesn’t allow for all of the server size features that are available, but for prototyping it is great. And when your application is ready to move up to a cluster size beyond M0, the GUI is there to make the transition rather pain free.
Atlas also provide for us, as users that care about our data and not just an easy interface, built in features such as security, replication, backups, automated upgrades, and fine-grained monitoring, to name a few. It allows for many of the painful tasks of database maintenance to be handled by MongoDB with the ability to provide alerts when things need more personal attention.
Of course there are lots of reasons why people don’t like the idea of storing their data in the cloud, outside their control, on someone else’s hardware. I’m not going to step into that argument as people seem to have strong opinions on the matter in both directions. However, if you are thinking of trying a cloud implementation of MongoDB, I think Atlas is definitely worth a look.
Originally published at Blog of Ken W. Alger.