As diversity and use cases of data are evolving rapidly, switching between different database management systems is becoming normal for organizations. 

If you are here, you might need to shift from MongoDB to Redshift. MongoDB is a non-relational database widely used by organizations to handle various data formats, high-speed data ingestion, and powerful query language. On the other hand, Redshift is a relational storage system used for its analytical capabilities, query performance, and storage capacity. Since both databases have different use cases, migration becomes an obvious choice for varying business requirements. 

This article will discuss migrating MongoDB to Redshift ETL using two easy methods. 

MongoDB Overview

Created in 2009, MongoDB is a modern NoSQL database. Being non-relational, it follows a document-oriented approach and stores data in documents that consist of key-value pairs and follow a BSON (Binary JavaScript Notation) structure. Unlike relational databases, MongoDB has no structural limitations. Therefore, it can process structured, semi-structured, and unstructured data. Additionally, its modern architecture and cutting-edge features like load balancing, query performance, and indexing make it stand out from other NoSQL databases.

Some of the key features of MongoDB include: 

• Schema-less Database: As a NoSQL database, MongoDB is schema-less, which means you can store different types of documents in one collection. In a MongoDB database, a single collection can hold multiple documents, and each document may consist of different numbers of fields and contents.‍
• Indexing: Indexes are unique data structures that store some information related to documents, making it easy for MongoDB to find the right data file. If the data is not indexed properly, the database searches each document with specified queries, which is time-consuming. ‍
• Rich Query Language: MongoDB has a rich query language for data querying and manipulation. You can use it to run searches for filtering data, sorting results, or even perform geospatial queries to locate data based on its coordinates. 

Redshift Overview

Amazon Redshift is a cloud data warehousing solution provided by Amazon Web Services (AWS). It can handle and process huge amounts of structured and unstructured data in the range of exabytes (10*18 bytes). Some of the well-known organizations that use Redshift in their tech stack include Amazon, Lyft, Figma, and Coursera. 

Key features of Redshift include:

• Petabyte Scale Data Warehouse: The distributed architecture of storage in Redshift supports workloads of up to 8 PB of compressed data. With this robust storage capacity, you can handle almost any number or type of nodes in your data warehouse. ‍
• Federated Query: The federated query feature of Redshift allows you to query live data across one or more databases without the need for data migration. This includes querying from Amazon's Relational Database Services (RDS), including Aurora MySQL Aurora PostgreSQL, and getting results as temporary tables.‍
• Serverless: Redshift offers a serverless architecture for data management. With this feature, you can manage any size of analytical workload without the need to manage a data warehouse's resources. Therefore, you can focus on the analytics rather than managing infrastructure.

Why Migrate From MongoDB to Redshift?

MongoDB and Redshift complement each other. While MongoDB is known for handling evolving and unstructured data, Redshift outshines it in analytical processing. Below are some of the key reasons for migrating MongoDB to Redshift: 

Data Warehousing

MongoDB: MongoDB is not a data warehouse. It offers a flexible NoSQL database that can handle diverse data types, including structured, semi-structured, and unstructured data.

Redshift: Redshift is a data warehouse. It lets organizations store data from one or more sources in a centralized repository. So that organizations can connect multiple databases of downstream applications into one data warehouse. This allows organizations to do cross-functional analysis and unify their data by creating a single source of truth. 

Business Intelligence & Reporting

MongoDB: MongoDB is ideal for handling transactional workloads and real-time applications. However, the analytical capabilities of MongoDB are not as robust as a dedicated data warehousing solution.

Redshift: Redshift streamlines the integration with well-known BI tools like Tableau and Power BI to extract meaningful insights from the organization's data. Organizations can use these BI capabilities to make data-driven decisions. 

Scalability

MongoDB: MongoDB provides horizontal scalability, which allows organizations to scale their database easily by adding more servers to a MongoDB cluster. 

Redshift: Redshift offers a Massively Parallel Processing (MPP) architecture to handle large datasets. Organizations can choose to add more computing resources to the Redshift cluster to scale up their analytics processing requirements. 

Methods to Sync MongoDB to Redshift

Method 1: Using Airbyte to connect MongoDB to Redshift ETL.

Method 2: Manually connecting MongoDB to Redshift. 

Method 1: Using Airbyte to Connect MongoDB to Redshift ETL

In this method, we use Airbyte, a widely used data integration tool, to extract data from MongoDB and load it into Redshift. Here is a step-by-step guide:

Step 1: Configure MongoDB As a Source

• Create an Airbyte account or Login. Navigate to the home page of Airbyte and click on the Sources button in the left navigation bar. 
• Use the search bar from the top on the Sources page and input MongoDB. Click on the connector card of MongoDB. 

• Then, you'll see a Create a Source page. Select your Cluster Type, and fill in other details, including the Connection String, Database name, Username, and Password of your MongoDB database. Toggle the advanced window and optionally make changes to default configurations. Click on Set up source.

Step 2: Configure Redshift As a Destination

• After setting up MongoDB as a source, click on Destinations from the left navigation menu. 
• On the Destinations page, select the search bar and type in Redshift. Click on the Redshift connector card. 

• You'll be redirected to the Create a destination section. Fill in the required details, including Host, Port, Username, Password, Database, and Uploading Schema. Click on Set up destination. 

Step 3: Connecting Source And Destination

• In the left navigation menu, click on Connections > Create a Connection. ‍
• Select MongoDB as a source and Redshift as a destination to establish a connection. 
• On the connection page, provide a unique Connection name and select the required Replication frequency. Optionally, you can tweak the Streams section and select your sync mode. 
• Click on Set up connection and run the sync once by clicking the Sync now button. 

That concludes it. You have now successfully connected MongoDB to Redshift. 

Method 2: Manually Connecting MongoDB to Redshift

In this method, we will manually migrate data from MongoDB to Redshift. It involves exporting data from MongoDB in JSON format, uploading it to the S3 bucket, and moving it to Redshift. Here's a detailed guide:

Prerequisites

• Amazon S3. 
• A SQL Client. 
• Windows Powershell. 

Step 1: Export Data From MongoDB

• Access your MongoDB database using the command prompt.

• Type the following command on the terminal to export data from your MongoDB collection in JSON format:

mongoexport --db db-name --collection collection-name --out output-file.json

Replace db-name with the name of the database, collection-name with the collection you want to export, and output-file.json with your desired file name. 

Step 2: Upload JSON file to S3 bucket

• Sign up or log in to the AWS Management Console and navigate to the Amazon S3 service. 
• Create a new bucket or use an existing one on S3.
• Upload the JSON file you exported from MongoDB on the S3 bucket. 

Step 3: Create a Table in Redshift

• Access your Redshift cluster using a SQL client. 
• Now, create a table schema in Redshift that matches the structure of MongoDB data in a JSON file. Type the following code in the SQL client: 

CREATE TABLE your_table( 
         column1 datatype1, 
         column2 datatype2,
          -- Add more columns as needed 
);

Replace your_table with the desired table name and define columns according to the JSON file structure. 

Step 4: Copy Data From S3 to Redshift

• On the SQL client, type the following code to copy data from the S3 bucket to the Redshift table: 

COPY your_table
FROM 's3_path' 
CREDENTIALS'aws_access_key_id=your_access_key_id;aws_secret_access_key=your_secret_access_key' 
FORMAT JSON' auto';

In the above code, replace your_table with your table name, s3_path with a path to your JSON file in S3, your_access_key_id, and your_secret_access_key with AWS credentials. 

Done. By following these steps carefully, you'll successfully export data from MongoDB, upload it on S3, and load it in Redshift. 

Conclusion 

Migration from MongoDB to Redshift can benefit organizations looking to enhance their analytical workloads and query performance. In this article, you have seen two popular methods for migrating data from MongoDB to Redshift. One method uses a SaaS tool - Airbyte, to streamline the process of migrating data. The other uses JSON files to migrate data manually. 

Both of these methods are widely used and have their use cases. However, using Airbyte, you can perform data migration within just a few clicks, and it automatically syncs the data if changes are made. However, in the manual method, you don't need a tool for migrating data, and the process is time-consuming and error-prone.

‍