This page provides you with instructions on how to extract data from Microsoft SQL Server and load it into Snowflake. (If this manual process sounds onerous, check out Stitch, which can do all the heavy lifting for you in just a few clicks.)
What is Microsoft SQL Server?
Microsoft SQL Server is a relational database management system that supports applications on a single machine, on a local area network, or across the web. SQL Server supports Microsoft's .NET framework out of the box, and integrates nicely into the Microsoft ecosystem.
What is Snowflake?
Snowflake is a cloud-based data warehouse that's fast, flexible, and easy to work with. It runs on Amazon Web Services EC2 and S3 instances, and separates compute and storage resources, enabling users to scale the two independently and pay only for resources used. Snowflake can natively load and optimize both structured and semi-structured data and make it available via SQL. It provides native support for JSON, Avro, XML, and Parquet data, and can provide access to the same data for multiple workgroups or workloads simultaneously with no contention roadblocks or performance degradation.
Getting data out of SQL Server
The most common way most folks who work with databases get their data is by using queries for extraction. With SELECT statements you can filter, sort, and limit the data you want to retrieve. If you need to export data in bulk, you can use Microsoft SQL Server Management Studio, which enables you to export entire tables and databases in formats like text, CSV, or SQL queries that can restore the database if run.
Preparing data for Snowflake
Depending on your data structures, you may need to prepare your data before loading. Check the supported data types for Snowflake and make sure that your data maps neatly to them.
Note that you won't need to define a schema in advance when loading JSON or XML data into Snowflake.
Loading data into Snowflake
Snowflake's documentation outlines a Data Loading Overview that can lead you through the task of loading your data. If you're not loading a lot of data, Snowflake's data loading wizard may be helpful, but for many organizations, its limitations make it unacceptable. Instead, you can:
- Use the PUT command to stage files.
- Use the COPY INTO table command to load prepared data into an awaiting table.
You can copy data from your local drive or from Amazon S3. Snowflake lets you make a virtual warehouse that can power the insertion process.
Keeping SQL Server data up to date
All set! You've written a script to move data from SQL Server into your data warehouse. But data freshness is one of the most important aspects of any analysis – what happens when you have new data that you need to add?
You could load the entire SQL Server database again. Doing this is almost guaranteed to be slow and painful, and cause all kinds of latency.
A better approach is to build your script to recognize new and updated records in the source database. Using an auto-incrementing field as a key is a great way to accomplish this. The key functions something like a bookmark, so your script can resume where it left off. When you've built in this functionality, you can set up your script as a cron job or continuous loop to get new data as it appears in SQL Server.
Other data warehouse options
Snowflake is great, but sometimes you need to optimize for different things when you're choosing a data warehouse. Some folks choose to go with Amazon Redshift, Google BigQuery, or PostgreSQL, which are RDBMSes that use similar SQL syntax, or Panoply, which works with Redshift instances. If you're interested in seeing the relevant steps for loading data into one of these platforms, check out To Redshift, To BigQuery, To Postgres, and To Panoply.
Easier and faster alternatives
If all this sounds a bit overwhelming, don’t be alarmed. If you have all the skills necessary to go through this process, chances are building and maintaining a script like this isn’t a very high-leverage use of your time.
Thankfully, products like Stitch were built to solve this problem automatically. With just a few clicks, Stitch starts extracting your Microsoft SQL Server data via the API, structuring it in a way that is optimized for analysis, and inserting that data into your Snowflake data warehouse.