Video Transcript (English)
Atlas is MongoDB’s database-as-a-service, which means that server configuration is largely handled for you. While more configuration options are available for self-managed deployments, there are some cluster settings that are configurable for an Atlas cluster.
In this video, you will learn how to access cluster and server configuration settings in the Atlas UI. You’ll also learn which server settings can be adjusted, and how to edit one of the most important settings of all: the number of nodes in a replica set.
Before adjusting settings for an Atlas cluster, you first need to consider: What is and isn’t possible to adjust on your current tier, the cost of the adjustments that you want to make, and the impact of the changes on your application.
We’ll keep these considerations in mind as we go through this lesson.
When you first got started with MongoDB Atlas, you probably created an M0 cluster. This is MongoDB’s free offering on a shared tier cluster. It’s a great option for exploring MongoDB Atlas!
A shared tier cluster is meant for smaller scale projects, so configuration options are limited. Most changes will require you to scale up.
Let’s talk more about your cluster configuration options.
Dedicated tier clusters offer options for development environments and low-traffic applications, as well as for high-traffic applications and large datasets.
The process for accessing and adjusting configuration settings for an M0 cluster all the way up to an M700 is the same. However, the settings that you are able to adjust depend on the tier.
If you choose to run MongoDB locally, you will have more configuration settings available to you.
To understand what’s possible at a larger scale in MongoDB Atlas, let’s explore some of the configuration options on an M20 dedicated tier cluster.
To access cluster configuration settings for any cluster, click here. Now we have access to Cloud Provider & Region, Cluster Tier, and Additional Settings. Server configuration settings are located under Additional Settings, so let’s start here.
Many of the settings that we see in this section have server parameter counterparts for a self-managed deployment.
For example, the Set Minimum Oplog Window affects the MongoDB server parameter, minRetentionHours.We can use this setting to specify the minimum number of hours to preserve an oplog entry. Require Indexes for All Queries affects the no table scan server parameter. We can activate this parameter to prevent queries that require a collection scan from running.
Notice here that there are fewer adjustable server parameters than in a self-managed deployment.
Let’s go back to our cluster configuration. Apart from changing the tier, cloud provider, and location of your deployment, you can also modify the number of nodes. In this section of the UI, we have access to more regions. We can enable Multi-Cloud, Multi-Region & Workload Isolation. When we enable this feature, we have the option to configure our deployment with 3, 5, or 7 electable nodes. In addition, we can add read-only nodes and analytics nodes depending on our application's needs. We can add a maximum of 50 nodes.
We can get fairly specific with our cluster configuration. For example: we can reconfigure our cluster to improve availability and local read performance in a specific region.
Here, we’re changing our node configuration to have two new nodes, in a new region, and to use two different cloud providers.
For most deployments, you won’t need this level of flexibility and high availability, but it is possible.
All done. Let’s recap what you learned in this video: In the Atlas UI, you can modify your cluster configuration to select free or paid tier, cloud provider, and hardware specifications. You can also adjust certain server settings for an Atlas dedicated cluster. These adjustable settings map to server parameters for a self-managed deployment. We can safely add and remove nodes from a replica set by using the Atlas UI, adjusting individual nodes’ level of priority, location, and cloud provider.