Unleash the Full Potential of Cast AI Platform with Demo Cluster

Table of Contents

1. Introduction
2. Exploring Cast AI Platform
   • Touring the Demo Cluster
   • Understanding the Overview of the Simulated Cluster
3. Onboarding a New Cluster and Savings Report
   • Optimizing Cluster Configuration
   • Comparison of Current and Optimized Cluster Configurations
   • Savings Calculation
4. Cost Monitoring and Workloads
   • Overview of Cluster Costs
   • Monitoring Individual Workloads
   • Tracking Costs by Namespace
   • Cost Comparison
5. Security Reporting
   • Best Practices and CIS Benchmarks
   • Vulnerability Scanning and Reporting
6. Auto Scaler Setup
   • Available Auto Scaler Configurations
7. Node Templates and Configurations
   • Creating Node Templates for Specific Workloads
   • Node Selector and GPU Node Configuration
   • Using Constraints and Default Configurations
8. Audit Log and Rebalancing
   • Tracking Actions in the Cast AI Cluster
   • Rebalancing and Optimization Process
9. Generating a Rebalance Plan
   • Comparing Existing and Resulting Cluster Costs
   • Blue-Green Deployment for Optimization
10. Conclusion

Introduction

In this article, we will Delve into the new demo cluster functionality on the Cast AI platform. If You are interested in exploring Cast AI and its optimization capabilities but do not want to onboard an existing cluster or lack a cluster to connect, the demo cluster built into the platform provides a solution. Join us as we take a tour and walk through the functionality of the Cast AI platform, discover different features, and explore the potential benefits it brings to your organization.

1. Exploring Cast AI Platform

1.1 Touring the Demo Cluster

The demo cluster on Cast AI allows users to navigate and experience the platform without connecting their own clusters. It provides a comprehensive overview of the simulated cluster, showcasing various functionalities and features. Through the guided tour, users can familiarize themselves with the platform and understand how it can optimize and streamline cluster management.

1.2 Understanding the Overview of the Simulated Cluster

When accessing the demo cluster, the user is presented with an overview of the simulated cluster. This screen displays essential information such as the number of nodes, the classification of instances (On-Demand, Fallback, or Spot), and the cluster's size. This data, in a real operational cluster, can be exported and utilized in external monitoring tools like Grafana or Prometheus.

2. Onboarding a New Cluster and Savings Report

Upon onboarding a new cluster on Cast AI, users are introduced to the available savings report, which plays a key role in optimizing cluster configurations and achieving cost savings. The report provides insights into the potential savings percentage by utilizing spot instances and optimizing the cluster's setup.

2.1 Optimizing Cluster Configuration

Cast AI analyzes the current cluster configuration and suggests an optimized configuration to maximize cost efficiency. By utilizing a combination of spot instances and On-Demand instances, significant savings can be achieved. The savings report presents the expected percentage of savings, considering an optimal spot instance usage.

2.2 Comparison of Current and Optimized Cluster Configurations

To Visualize the potential benefits, the savings report provides a side-by-side comparison of the current and optimized cluster configurations. Users can identify the differences between the two setups and assess the impact on cost optimization and performance.

2.3 Savings Calculation

The savings report quantifies the potential cost savings by comparing the current and optimized cluster configurations. It highlights the potential cost reduction in terms of monthly expenditures. Users can understand the financial benefits that can be anticipated by implementing Cast AI's optimization strategies.

...