# AWS Redshift Cluster Calculator AI Agent Connect

> Deterministic sizing for AWS Redshift clusters, including node counts, concurrency scaling, and WLM configuration.

## Overview
- **Category:** cloud-infrastructure
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_OJIVXy1WREFPoEekc6MBLkCmNskfCzYlKcrSevSj/ai-agent-connect
- **Tags:** aws, redshift, data-warehouse, infrastructure, sizing

## Description

This MCP server provides precise sizing and configuration for AWS Redshift clusters. Use `calculate_cluster_size` to determine the required number of nodes based on data volume and hardware profiles like dc2, ds2, or ra3. Optimize your environment using `optimize_workload_performance` to manage concurrency scaling slots, distribution styles, and sort key strategies. Finally, use `configure_resource_management` to set up WLM queues and snapshot schedules. It acts as a bridge between your AI agent and AWS Redshift architecture planning.

## Tools

### calculate_cluster_size
Determines the fundamental hardware requirements for the Redshift cluster

### configure_resource_management
Sets up the internal governance of the cluster, including memory allocation and data backups

### optimize_workload_performance
Provides configuration for query execution, scaling, and data organization

## Prompt Examples

**Prompt:** 
```
How many nodes do I need for 50TB of data using dc2 nodes?
```

**Response:** 
```
For 50TB of data on dc2 nodes, you will need a specific number of nodes based on the storage capacity of the dc2 tier. Please run `calculate_cluster_size` with 50 as the data volume to get the exact count.
```

**Prompt:** 
```
Help me optimize a cluster with 5 nodes and 20 concurrent queries.
```

**Response:** 
```
To optimize a 5-node cluster with 20 concurrent queries, use `optimize_workload_performance`. This will calculate available concurrency scaling slots and recommend a distribution style like 'even' or 'key' to maximize performance.
```

**Prompt:** 
```
Set up a snapshot schedule for a 10-node cluster.
```

**Response:** 
```
You can configure the snapshot schedule by calling `configure_resource_management` with 10 nodes and your preferred frequency (hourly, daily, or weekly).
```

## Frequently Asked Questions

**How do I determine the number of nodes needed?**
You can use the `calculate_cluster_size` tool. Provide the total compressed data volume in TB and select your preferred node type (dc2, ds2, or ra3) to get the exact node count.

**Can I configure WLM queues through this tool?**
Yes, the `configure_resource_management` tool allows you to define up to 8 WLM queues and allocate memory percentages to ensure efficient workload management.

**What is the difference between compound and interleaved sort keys?**
When using `optimize_workload_performance`, you can choose between compound sort keys (prioritized sequence) or interleaved sort keys (multi-dimensional weight) depending on your query patterns.