Fivetran with its ELT process enables us to move the data from any source to destination. With Fivetran, no engineering required, it will take minutes with a few clicks and not months to build a pipeline.
With modern cloud data warehouse like snowflake getting very popular with enterprises – they are leveraging tools like Fivetran for real-time replication of data into the snowflake and/or other cloud data warehouse.
Fivetran has a real-time dashboard to monitor the Fivetran performance, Fivetran status. Below is a snapshot of the system’s status provided by Fivetran for monitoring the entire Fivetran’s status.
Does Fivetran has its own warehouse to store the data, or does it have any specific source that cannot be used with Fivetran? Will Fivetran load my historical data too? The answer to these questions and few more will be discussed in detail in this blog.
Introduction to connectors
With No-coding guaranteed, Fivetran has hundreds of connectors to push or pull your data from your source and load it to your destination cloud data warehouse with 5 minutes of setup.
Upon creating a single Fivetran account you can use different connectors for different sources like Applications, Databases, events, Files, and cloud functions, and copy data into a single destination of your choice. These connectors can be paused, Re-synced, or deleted based on your requirement. Destination schema + table name have to be unique for each connector. If the destination schema is not available Fivetran will create a schema automatically, then create the table and load it.
Let us have a quick peek at the important things that we might want to consider while planning to move to Fivetran.
1. Do not have a destination?
Fivetran got that coved, it provides a free Fivetran supported destination, BigQuery data warehouse. The warehouse is free to use for your trial period of 14 days. Your 14-day free trial begins once your initial sync is finished.
2. Can I have a completion date for the Fivetran implementation?
Since the duration of this process varies for each customer completion date cannot be fixed for initial sync. Your 14 days of the trail starts after completion of your initial sync so either wait for the completion of the initial sync or include an opt-out clause in your contract that dissolves the agreement if the initial sync is not completed within a specific time frame.
3. Can I skip historical syncs? Or sync from a specific date?
No, you cannot skip your historical sync or start your sync from a specific date (for example, only sync data from August 2014 forward).
4. Does Fivetran support One-time production data migrations?
Fivetran does not support one-time migrations of production data.
5. Can data warehouse without change capture mechanism be a source?
No, you cannot connect data warehouse like Teradata, Netezza, or Vertica as a source, because these have no change capture mechanism.
6. Excluding Source data, can selected tables move to the destination?
You can easily exclude schemas, tables, or columns from your syncs on your Fivetran dashboard. Alternately, you can change the permissions of the Fivetran user in your source database and restrict its access to certain tables or schemas. If your source is a file system, a single storage bucket, and its underlying data is loaded into a single destination table. Any files added to that bucket under the specified folder will be pulled and loaded into the specified destination table.
These are the widely discussed items on Fivetran. With Fivetran, we can create a pipeline that will load the data from any source to a destination within a few minutes. Fivetran also provides us with options to append or update/insert records while moving from source to destination.
Fivetran is very strong for its real-time extraction and loading capabilities (EL) and is rapidly improving it’s data transformational capabilities as well so that it can be a strong ETL offering for customers looking for real-time replication to their cloud data warehouses like Snowflake.