What does this Block do for me?

(1) Reduce Costs and Optimize Performance - Gain a holistic view across your datawarehouse and optimize across the board. Identify your longest running queries, most costly queries, troublesome users, and more, to uncover the most impactful cost-saving and performance-enhancing opportunities.

(2) Monitor Cost and Usage Data - Provide comprehensive insights around your Snowflake deployment and usage across all your tables, databases and users. See how these metrics trend over time and compare to prior periods to ensure you're scaling effectively.

(3) Compare Performance Across Databases - If you're using another database, in addition to Snowflake, and want to understand the cost and performance implications for your workloads across your cloud environments, compare the Snowflake results from this Block with the results from other databases, such as Redshift (Block here) or BigQuery(Block here).

(4) Enterprise Data Platform - Take advantage of Looker's data platform functionality, including data actions, scheduling, permissions, alerting, parameterization (each user can only see their own data), and more. Get immediate alerts when usage is exceeded, queries exceed a runtime threshold, or any other business criteria.

(5) Understand Data Loading Issues - Analyze data loading successes and failures into any of your snowflake databases and track or alert on unexpected changes.

Snowflake Account Usage Data Structure

• Snowflake's Account Usage dataset comes complete with several tables (called "views"), such as databases, functions, query history, and much more. Colectively these tables give a comprehensive overview of Snowflake operations, execution steps, and processing time.

Block Structure

• This Block is built on the ACCOUNT_USAGE share provided by Snowflake (using their Sharehouse offering).
• Each Snowflake table is represented as it's own view in Looker. The Model file of this Block joins together the underlying views based on available keys. It provides a starting place for additional custom modeling. This Block covers all of the core metrics that come with the dataset, as well as additional value-add analysis, and should serve as a great jump start for further exploration.
• The schema documentation for Account Usage can be found in Snowflake's documentation. Please note that your naming might vary slightly.

Additional Info

Data Latency

• The INFORMATION_SCHEMA views and table functions display data in real-time, whereas the ACCOUNT_USAGE views have some built-in latency, due to the process of extracting the usage data from Snowflake’s internal metadata store. However, the ACCOUNT_USAGE views have been designed to minimize the latency impact, based on the type of data displayed in the views:

Runtime:

• Approximately 1 minute for the following views: QUERY_HISTORY, LOGIN_HISTORY

Analytic:

• 1 hour (or less) for the following views: WAREHOUSE_METERING_HISTORY, STORAGE_USAGE, LOAD_HISTORY, and DATABASE_STORAGE_USAGE_HISTORY

Dictionary:

• 1 hour (or less) for the following Columns: DATABASES, FILE_FORMATS, FUNCTIONS, REFERENTIAL_CONSTRAINTS, SCHEMATA, SEQUENCES, STAGES, TABLE_CONSTRAINTS, TABLE_STORAGE_METRICS,TABLES, and VIEWS

Implementation Instructions

Accessing the ACCOUNT USAGE Share

To access the ACCOUNT_USAGE share, an account administrator must grant the user provided in your looker connection with access to the appropriate schema.

As an example, The commands below would be used if the user in your looker connection was granted the "looker_role" and that is how you plan on permissioning the SNOWFLAKE shared DB:

grant imported privileges on database SNOWFLAKE to role looker_role;
grant select on all views in schema SNOWFLAKE.ACCOUNT_USAGE to role looker_role;

Dashboards

The dashboard can be used as is, or customized to your specific requirements. If you've customized the model name, rename the model in each LookML Dashboard element from "snowflake_usage_block" to the model name you've selected. We recommend using a global Find & Replace for this.

What if I find an error? Suggestions for improvements?

Great! Blocks were designed for continuous improvement through the help of the entire Looker community, and we'd love your input. To log an error or improvement recommentation, simply create a "New Issue" in the corresponding Github repo for this Block. Please be as detailed as possible in your explanation, and we'll address it as quick as we can.