> ## Documentation Index
> Fetch the complete documentation index at: https://docs.plumi.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Snowflake

> Connect to Snowflake cloud data platform

## Overview

Connect Plumi to Snowflake for cloud data platform analytics with automatic scaling and near-unlimited concurrency.

## Connection Details

| Field     | Description                  | Example             |
| --------- | ---------------------------- | ------------------- |
| Account   | Snowflake account identifier | `xy12345.us-east-1` |
| Warehouse | Virtual warehouse name       | `COMPUTE_WH`        |
| Database  | Database name                | `ANALYTICS`         |
| Schema    | Schema name (optional)       | `PUBLIC`            |
| Username  | Snowflake user               | `PLUMI_USER`        |
| Password  | User password                | `********`          |

## Account Identifier

Your account identifier includes:

* Account name
* Region (if not in US West)
* Cloud provider (if not AWS)

Examples:

* `xy12345` (US West on AWS)
* `xy12345.us-east-1` (US East on AWS)
* `xy12345.us-east-1.azure` (US East on Azure)

## Setting Up a Read-Only User

Create a dedicated read-only user:

```sql theme={null}
-- Create role
CREATE ROLE PLUMI_READONLY;

-- Grant warehouse usage
GRANT USAGE ON WAREHOUSE COMPUTE_WH TO ROLE PLUMI_READONLY;

-- Grant database access
GRANT USAGE ON DATABASE ANALYTICS TO ROLE PLUMI_READONLY;
GRANT USAGE ON ALL SCHEMAS IN DATABASE ANALYTICS TO ROLE PLUMI_READONLY;
GRANT SELECT ON ALL TABLES IN DATABASE ANALYTICS TO ROLE PLUMI_READONLY;

-- Grant future table access
GRANT SELECT ON FUTURE TABLES IN DATABASE ANALYTICS TO ROLE PLUMI_READONLY;

-- Create user
CREATE USER PLUMI_USER
  PASSWORD = 'your_secure_password'
  DEFAULT_ROLE = PLUMI_READONLY
  DEFAULT_WAREHOUSE = COMPUTE_WH;

-- Assign role to user
GRANT ROLE PLUMI_READONLY TO USER PLUMI_USER;
```

## Query Syntax

Snowflake uses ANSI SQL with extensions:

```sql theme={null}
-- Date functions
SELECT
  DATE_TRUNC('month', order_date) as month,
  SUM(total) as revenue
FROM orders
WHERE order_date >= '2024-01-01'
GROUP BY month;

-- Semi-structured data (JSON)
SELECT
  data:customer:name::string as customer_name,
  data:order:total::number as total
FROM orders_json;
```

## Cost Optimization

Snowflake charges by compute time. To minimize costs:

* Use appropriate warehouse sizes
* Suspend warehouses when not in use
* Use result caching (automatic for repeated queries)
* Cluster tables on frequently filtered columns

## Troubleshooting

### "Account not found"

Verify your account identifier includes the correct region and cloud provider.

### "Warehouse suspended"

The warehouse may be suspended. Plumi will attempt to resume it, but ensure the user has `OPERATE` privilege on the warehouse.