# Snowflake Cost Optimization

The **Snowflake Cost Optimization** dashboard, part of Paradime's Radar suite, allows you to control AI-driven cost-saving features that automatically reduce costs while maintaining performance. This section offers two powerful tools designed to maximize your Snowflake efficiency: **Warehouse AI Agent Optimizer** and **Warehouse AI Autoscaler**. 

{% hint style="warning" %}
**Prerequisites**

* Completed the [**Cost Management**](https://docs.paradime.io/app-help/documentation/getting-started#id-1.-cost-management) setup in Radar's [Get Started guide](https://docs.paradime.io/app-help/documentation/radar/getting-started).
  {% endhint %}

### Key Features

{% tabs %}
{% tab title="🔄  Warehouse Autoscaler" %}

* 24/7 usage analysis and automatic rightsizing
* Adapts to time-of-day and utilization patterns
* Conservative approach maintains query performance
* Optimized for BI tools and reporting warehouses
  {% endtab %}

{% tab title="⏳ Idle Time Optimizer" %}

* AI-powered learning from warehouse monitoring
* Reduces wasted idle time
* Optimizes suspension timing across all executions
* Works seamlessly with your existing Snowflake setup
  {% endtab %}
  {% endtabs %}

{% @arcade/embed flowId="ydGGg7gNr73z2y78YrTR" url="<https://app.arcade.software/share/ydGGg7gNr73z2y78YrTR>" %}

***

### Savings Overview

Value: Track your potential annual savings, actual savings to date, and review how optimizations have impacted your warehouse spend.

<figure><img src="https://2337193041-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHET0AD04uHMgdeLAjptq%2Fuploads%2FzlQ4Z5keZafWx1xeklT0%2Fimage.png?alt=media&#x26;token=700aa3de-8bad-473b-998d-40a9e8cab773" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
**How to Use**:

1. **Review savings**:\
   Check **potential annual savings** and **total savings to date** using the **Warehouse Optimizations** table. Apply the **Date Range Filter** (7, 14, 30, or 60 days) to focus on specific periods.
2. **Monitor the savings chart**:\
   Use the **Warehouse Savings Across Optimizations** chart to track optimization performance. Apply the **Optimization Type Filter** to focus on specific optimizations like the **AI Agent** or **Autoscaler**.
3. **Adjust strategy**:\
   Review cost-saving trends with the **Date Range Filter** and adjust optimizations accordingly to maximize savings over different periods.
   {% endhint %}

***

### Warehouse Optimizations

You can enable or disable these optimizations based on your cost-saving goals while maintaining performance. The two available optimization options are:

**1. Warehouse AI Agent Optimizer**

The **Warehouse AI Agent Optimizer** is responsible for optimizing the configuration of your Snowflake warehouses based on real-time and historical usage data. It ensures that your warehouses are right-sized and configured optimally without manual intervention, helping you save costs while maintaining query performance.

<figure><img src="https://2337193041-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHET0AD04uHMgdeLAjptq%2Fuploads%2FjrhkShcBUSlw0PUS7v2D%2Fimage.png?alt=media&#x26;token=370c95cb-1586-452b-be3e-8b12498506f8" alt=""><figcaption></figcaption></figure>

**How it Works**:

* The AI agent reviews your warehouse usage patterns and automatically adjusts configuration settings to optimize performance.
* The optimizer reduces the resources allocated to underused warehouses while maintaining adequate capacity for peak usage times.

{% hint style="info" %}
**How to Use**:

* **Enable the Warehouse AI Agent Optimizer** for specific warehouses using the **Warehouse Optimizations** table.
* **Monitor potential savings** for each warehouse to see where the most savings can be generated.
* **Track actual savings** using the "Savings" column in the table and adjust optimizations as needed based on performance.
  {% endhint %}

***

#### 2. Warehouse AI Autoscaler

The **Warehouse AI Autoscaler** dynamically adjusts the size of your Snowflake warehouses based on the workload. It scales resources up or down automatically based on demand, ensuring that you're only using the capacity you need, which reduces costs while maintaining performance.

<figure><img src="https://2337193041-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHET0AD04uHMgdeLAjptq%2Fuploads%2FePSvG2qSReeTsAAasbEq%2Fimage.png?alt=media&#x26;token=4a73428b-ba71-4a4e-bcbd-4988284def0b" alt=""><figcaption></figcaption></figure>

**How it Works**:

* The Autoscaler monitors the real-time query workload for each warehouse and automatically resizes it.
* It ensures resources are right-sized during peak demand while scaling down during off-peak hours to prevent unnecessary resource costs.
* The Autoscaler is optimized for BI tools and reporting warehouses, making it ideal for environments with fluctuating workloads.

{% hint style="info" %}
**How to Use**:

* **Enable the Warehouse AI Autoscaler** for specific warehouses using the **Warehouse Optimizations** table.
* Review the estimated savings generated by the Autoscaler and compare it to your overall warehouse costs.
* Use the savings data to ensure that warehouses with fluctuating demand are automatically right-sized to minimize costs.listed.
  {% endhint %}

***

{% hint style="success" %}

#### Best Practices:

* **Start with Your Most Expensive Warehouses**:\
  Focus your efforts on the warehouses with the highest potential savings, as these can generate the largest cost reductions.
* **Regularly Review Optimization Performance**:\
  Use different time ranges (7, 14, 30, or 60 days) to analyze how well your warehouses are being optimized. Adjust settings if you notice savings aren't as high as expected.
* **Pay Attention to Warehouses with High Potential but No Current Savings**:\
  Review warehouses that show high potential savings but currently have no savings. This may indicate they need further adjustments or more aggressive optimizations.
  {% endhint %}