What to know when working with Google’s Data Warehouse
Best Practices when working with Google’s BigQuery
BigQuery is a SaaS (Software as a Service) Data Warehouse technology and will charge you in a pay for what you use pricing model. In this short article, I want to show you what the prices are and how you can optimize your use.
Which Actions do actually cost money?
The good thing first — a few actions within BigQuery are for free, like:
- Load Data (From a Bucket e.g)
- Extract Data (To a Google Sheet e.g)
- Copy Data (To a new table e.g)
- Delete Tables, Views and Datasets
- Cluster and Partition Tables
But you can probably guess that the most common activities such as storage and querying do cost some money. Especially the last function is the one that Google uses for earning money. For more details visit Googles BigQuery pricing .
How much do the Actions cost?
The common overview will give you a first insight of what the costs of the most common tasks in BigQuery look like:
- Query: charged by bytes processed will be $5 per TB (first TB each month is free)
- Storage: charged by GB stored per month will be $0.02 (Active Storage) and $1.10 per TB for the BigQuery Storage API
- Streaming Inserts will cost you $0.05 per GB
These are the current prices shown by Google for the US region, other regions differ in price.
How is BigQuery compared to other Technologies?
Comparing prices to other technologies like Amazon’s Redshift or Snowflake is difficult because it depends on the use case and how you optimize each technology. Also, the prices are very similar — here fivetran has released a very useful benchmark . However, it is important to emphasize that BigQuery offers the highest SaaS feeling — so you don’t have to worry as much about provisioning and operations, which again has a positive impact on costs…