> For the complete documentation index, see [llms.txt](https://docs.akamas.io/insights/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.akamas.io/insights/configuration/settings-configuration.md).

# Settings

The Settings page allows you to customize how Akamas Vivo calculates costs and generates recommendations. Navigate to **Recommendations** > **Settings** from the main menu.

![Settings Page](https://987903283-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FAxUCNi77PcMdZAmajVK6%2Fuploads%2Fgit-blob-542b70d0d1468334c416b1a72f205f6366c9c108%2Fsettings-page.png?alt=media)

## Instance Catalog

Costs are calculated from the instance catalog that resolves for each cluster: the provider's public price list, a discount on it, or a custom catalog you upload. Manage them on the **Instance Catalog** tab of the Settings page.

### Fallback pricing

A cluster no catalog resolves for, such as an on-premise cluster with no catalog assigned, is priced per CPU core-hour and per GiB-hour from the **Fallback pricing** row at the bottom of the catalog list.

The two rates matter on every cluster too, through their ratio: it sets how a workload's cost splits between CPU and memory, and weights the cost-based health scores. Changing the ratio therefore moves workload costs and scores on clusters priced from a catalog, while their cluster totals stay those of the catalog.

* **CPU price**: Cost per core per hour ($/cores/hour). Example: `0.0416` for approximately $30/core/month.
* **Memory price**: Cost per GiB per hour ($/GiB/hour). Example: `0.0056` for approximately $4/GiB/month.

The platform values (CPU `0.04048` $/core/hour, memory `0.004445` $/GiB/hour) apply until you set your own; **Reset to platform values** restores both. Each rate must be above zero and at most `100` per hour. Clusters priced this way are listed on the same panel when a data import is selected. If the row shows **Unavailable**, the pricing service could not answer; the catalogs above it stay usable, and costs use the last rates read, or the platform values if none were read yet, until it recovers.

Label-based custom price rules are retired: to price a group of clusters differently, assign them a catalog of their own on this tab.

## Optimization Profiles

Optimization profiles control how aggressively Akamas Vivo recommends resource changes. Each profile adjusts safety margins and utilization targets differently.

### Available Profiles

* **Conservative**: Larger **safety** margins, lower utilization targets. Prioritizes reliability over savings. Best for critical production workloads.
* **Balanced**: Moderate safety margins and utilization targets. Good **mix** of **savings** and risk **mitigation**. Suitable for most workloads.
* **Aggressive**: Smaller safety margins, higher utilization targets. Maximizes **cost savings**. Appropriate for fault-tolerant workloads.

### Selecting a Profile

Use the **Profile Selector** dropdown available throughout the application. All recommendations and metrics update immediately when you change profiles.

{% hint style="info" %}
Profile changes affect all pages in the application, not just Settings.
{% endhint %}

### Custom Profiles

For assistance creating custom profiles tailored to your specific requirements, contact your Akamas representative.
