Turn Your Ops Team Into MongoDB Believers

MongoDB adoption has exploded in the last few years, making it the leading NoSQL database. This widespread acceptance is largely due to its appeal among developers – MongoDB has many traits that make it the go-to option to rapidly develop applications.

While developers are using MongoDB in dev and test environments, these apps must also be supported by the Ops team in order for them to be deployed to production. Therein lies the challenge. MongoDB is a new technology—it’s relatively immature and many operations teams are unfamiliar with it. Getting your Ops team on board with MongoDB can be tricky.

Here are some ways you can showcase the advantages of MongoDB to your Ops team as you look to use it as the database for a production app or site. These talking points can help you turn your operations team into MongoDB believers.

Speak to Maximizing Uptime
“An Ops engineer wants something to run 24/7. Being performant is good, but they want it to be up,” Major Hayden, Rackspace Chief Security Architect, advises. One key feature of MongoDB is that it is schemaless, meaning there are no more migrations due to schema changes. This can be a dramatic boost for uptime. Hayden further advises developers to discuss the redundancy features and three-node replica sets that can help increase availability.

Numbers Talk: Provide Benchmarking
Ops engineers are keen on data and numbers, so show them how MongoDB can outperform the current relational database with benchmarking. “If you can prove out your use case for MongoDB with hard numbers, there is hardly any room for arguing against using it,” Rackspace operations engineer Farid Saad says.

Reliability and Scale Testing
“As someone with a strong background in relational databases, I know exactly how a MySQL database or cluster would respond if you turned off a server,” Rackspace operations engineer Richard Maynard says. “But in the MongoDB world, I don’t have the same insight and I feel many Ops engineers are in a similar position.”

Maynard suggests working with the Ops team to perform reliability and scale testing to demonstrate the reliability of MongoDB. By simulating the amount of data you expect to accumulate at the end of one year, the operations team can look at how the system performs, where possible issues may exist and even what happens if a server fails. This helps convert that fear of the unknown into a known response.

Demonstrate the Ease of Configuration
Show your Ops team how MongoDB can work seamlessly with configuration management tools like Chef, Puppet or Saltstack. Ops engineers love it when there is less manual configuration involved or fewer manual workarounds to deploy, so demonstrate how this automation can be applied to MongoDB. “To get an Ops engineer to use MongoDB, show them how easy it is to configure and maintain that configuration,” Rackspace Deployment Services engineer BK Box advises.

Speak to Available Hosting Partners
While your Ops team may not be familiar with MongoDB, there are hosting providers who specialize in this NoSQL database. ObjectRocket is supported by MongoDB experts and is built on fine-tuned hardware to get the best performance out of the database.

“Many customers come to ObjectRocket because their previous provider has no technical understanding of MongoDB. They need more help and many times just don’t want to be in the business of managing the database in production,” says Kenny Gorman, a co-founder of ObjectRocket by Rackspace. “We help them get the most out of MongoDB.” With ObjectRocket, your MonogDB server is fully sharded and replicated, and it has the ability to scale in a few clicks. This can help free up your Ops team to focus on administering the web and application tier while the specialists handle the MongoDB database.

Looking to implement MongoDB without having to worry about the backend? Learn more about how ObjectRocket fine-tunes hardware to get the highest performance out of MongoDB. 


  1. The one thing standing in the way of my team embracing Mongo DB is the lack of admin tools. Relational database have an abundance of GUI and other tools that allow ad hoc data gathering and manipulation. There are very few for Mongo, and those that exist expand very little on Mongo’s unfamiliar CLI syntax. This is a problem in both development and production.

  2. Very nice read Garrett, thanks for sharing. I am a counselor for Hadoop beginners. Am in the process of putting together a document with few tips from Industry professionals, as in what is required to get started with Hadoop. Was wondering if you could help me out with a few tips of your own for Hadoop freshers.

  3. Is this a joke? MongoDB is one of the biggest jokes in the storage business…it’s only fast because it make you do all the hard stiff yourself…like synchronization, reliable replication and rolling back. If you use MongoDB, then you too can have your own war stories…


Please enter your comment!
Please enter your name here