Breadcrumbs

Cloudamize Express AMD Designs

Overview

When reviewing your finalized Cloudamize Express flat-file analysis, the platform provides dedicated algorithmic modeling tracks designed to identify target instance families backed by high-efficiency AMD processors.

By leveraging the Design dropdown menu located at the top of your interactive center panel workspace, you can instantly toggle and filter your assessment results across four automated AMD strategy paths. These strategies evaluate hardware inventory inputs to match your unique operational goals, whether you are looking for zero-overhead migrations, maximum infrastructure cost-cutting, or optimized computational throughput.

AMD Cloud Sizing & Optimization Models

The Cloudamize Express engine utilizes four precise optimization profiles to map flat-file inventory details directly to Amazon Web Services (AWS) infrastructure specifications:

1. AMD Easy Path

  • Strategic Intent: Low-risk, predictable migration mapping.

  • How it Works: This model defaults to recommending AMD's 6th generation instances (with most recommendations resolving to 6a instances) for all workloads. If an instance type in the source flat file is already at a generation above the 6th, it will remain as is with no changes applied.

2. AMD Performance Optimized

  • Strategic Intent: Strict compute stability with enhanced processing capability.

  • How it Works: This model recommends the specific AMD machine that provides superior performance compared to the source instance type without changing the shape of the machine. The allocated vCPU and Memory configurations are kept exactly the same. Most recommendations in this plan will resolve to 8a instances to capture newer processor performance advantages.

  • Example: An Intel-based r6i.24xlarge (96 vCPUs, 768 GB Memory) is mapped to a high-performance AMD r8a.24xlarge (96 vCPUs, 768 GB Memory), retaining the exact resource sizing footprint while maximizing performance throughput.

3. AMD Cost Optimized

  • Strategic Intent: Structural infrastructure cost-cutting through performance-aware sizing.

  • How it Works: This model looks for a target machine that delivers higher performance than the source while sitting at a lower cost bracket. Under this logic, the engine does not retain the shape of the machine, but it preserves the instance family tier. It may downsize the instance layout if it finds an alternative AMD machine that outperforms the source while maximizing budget efficiency.

  • Example: A source instance configuration like an r6i.24xlarge can be optimized and downsized to an r8a.8xlarge under this logic.

4. AMD Optimized

  • Strategic Intent: The master recommended hybrid plan.

  • How it Works: Rather than applying a single uniform rule across your entire flat file, this model builds a single hybrid plan by executing a precise, sequential evaluation matrix. For every parsed server node, it sweeps the top three plans in a specific order of precedence:

    1. First Check: Evaluates the workload under the Easy Path rules. If a valid, optimized recommendation is found, it selects it.

    2. Second Check: If the workload cannot be optimized via Easy Path, it drops down to be evaluated under the Cost Optimized framework.

    3. Third Check: If no fit is found in the previous tiers, it defaults to the Performance Optimized track.

Comparing Sizing Results Across Designs

You can use the header design picker dropdown on your Express Dashboard to switch between these tracks in real time. Toggling between these profiles will instantly recalculate the core metrics inside your recommendation table:

  • Annual Cost Deltas: See how your net percentage savings (e.g., ↓ 31.8% Savings) shift based on a Cost vs. Performance strategy.

  • Hardware Sizing Specs: Review the side-by-side changes to Recommended Instance Type, Recommended vCPU, and Recommended Memory (GB) as the algorithmic threshold adjusts.

  • Exporting Options: Once you select a preferred track, click Download Summary Deck to download an executive presentation of that specific design, or click Export to save the raw multi-column CSV dataset.

    If you have any queries, please get in touch with the helpdesk via our Helpdesk Portal or by email at helpdesk@cloudamize.com.