Overview
This article provides a detailed analysis of how the five major cloud data warehouses—Snowflake, Databricks, ClickHouse Cloud, Google BigQuery, and Amazon Redshift Serverless—calculate compute costs. It emphasizes the importance of understanding billing models over price lists to accurately compare performance and costs for analytical workloads.
What You'll Learn
1
How to compare compute costs across different cloud data warehouses
2
Why understanding billing models is crucial for cost-effective cloud usage
3
How to utilize the Bench2Cost framework for cost analysis
Key Questions Answered
How do the five major cloud data warehouses bill for compute?
Each cloud data warehouse has a unique billing model based on different units like credits, DBUs, compute units, slot-hours, and RPUs. Understanding these models is essential to accurately compare costs and performance across platforms.
What is the Bench2Cost framework and how does it work?
The Bench2Cost framework translates benchmark runtimes into dollar costs by mapping raw query execution times to the specific billing models of each cloud data warehouse. It produces an enriched dataset that allows for direct cost comparisons.
What are the key differences in compute models among these cloud data warehouses?
Snowflake and Databricks use provisioned clusters, ClickHouse Cloud offers flexible compute units, while BigQuery and Redshift Serverless utilize serverless models. Each model affects how costs are calculated and billed.
Key Statistics & Figures
Snowflake pricing per credit
$2.00 / credit / hour
Standard pricing for Snowflake in AWS, US East as of November 2025.
Databricks pricing per DBU
$0.70 / DBU / hour
Pricing for Databricks SQL Serverless in AWS, US East as of November 2025.
ClickHouse Cloud pricing per compute unit
$0.22 / unit / hour
Basic tier pricing for ClickHouse Cloud in AWS, US East as of November 2025.
BigQuery on-demand pricing
$6.25 / TiB scanned
On-demand pricing for BigQuery in US-East as of November 2025.
Redshift Serverless pricing per RPU
$0.36 / RPU / hour
Pricing for Redshift Serverless in AWS, US East as of November 2025.
Technologies & Tools
Some links below are affiliate links. We may earn a commission if you make a purchase.
Data Warehouse
Snowflake
Used for analytical workloads with a provisioned compute model.
Data Warehouse
Databricks
Provides SQL Serverless capabilities for data analytics.
Data Warehouse
Clickhouse Cloud
Offers flexible compute units for scalable analytics.
Data Warehouse
Google Bigquery
Utilizes a serverless shared compute model for query execution.
Data Warehouse
Amazon Redshift Serverless
Employs serverless RPUs for dynamic scaling and cost management.
Key Actionable Insights
1Understand the billing units of each cloud data warehouse to make informed decisions.Different cloud data warehouses use various billing units like credits or DBUs, which can significantly impact your costs. Knowing these details helps in budgeting and optimizing resource allocation.
2Utilize the Bench2Cost framework for accurate cost comparisons.By applying the Bench2Cost framework, you can translate benchmark results into comparable cost metrics, enabling you to choose the most cost-effective solution for your analytical workloads.
3Consider the scaling capabilities of each cloud data warehouse when planning workloads.Understanding how each system scales—whether vertically or horizontally—can help you optimize performance and manage costs effectively, especially during peak usage times.
Common Pitfalls
1
Relying solely on price lists for cost comparisons can be misleading.
Price lists do not reflect the actual compute costs due to varying billing units and execution models. Understanding the underlying billing mechanisms is essential for accurate cost assessments.