Benefits of Using Wayfinder’s Compute Templates for Self-Service Cluster Provisioning

As a developer, you want to spend your time building and improving apps, not worrying about the details of cloud infrastructure. But when deploying apps across multiple clouds like AWS, Azure, or GCP, it’s important to choose the right compute instances - without wasting time or resources.

‍

That’s where Wayfinder’s compute templates can help. These templates simplify the provisioning process, making it faster and easier to select the right compute instances without needing to manually handle cost calculations. They’re designed to give you flexibility while staying within budget, thanks to built-in guardrails that prevent overspending.

‍

What are Compute Instances?

Compute instances are the backbone of your cloud infrastructure. In cloud environments like AWS, Azure, or GCP, these virtual machines (VMs) run your applications. They can be customised to match your specific needs - whether you require more CPU, memory, or storage.

‍

Choosing the right instance type directly affects your workload’s performance and cost. For example, if your app is handling basic database requests, a general-purpose instance might be all you need; there’s no need to overprovision with a high-performance option.

‍

Here’s a quick overview of the different compute instance types available across AWS, Azure, and GCP:

‍

• General Purpose: Balanced options for general tasks like web servers or small databases. Covers: AWS’s T3, M5, A1 series; Azure’s D-series, B-series; and GCP’s N1, N2, E2 series.
• Compute Optimised: Ideal for compute-heavy tasks like scientific modelling.
  Covers: AWS’s C5 and C6g series; Azure’s F-series; and GCP’s C2 series.
• Memory Optimised: Perfect for high-performance databases or big data analytics. Covers: AWS’s R5, X1e series; Azure’s E-series, M-series; and GCP’s M1, M2 series.
• Storage Optimised: For high-speed read/write needs. Covers: AWS’s I3, D2 series; Azure’s Lsv2-series; and GCP’s N2-highmem, N2d-highmem series.
• Accelerated/GPU Compute: Great for machine learning or video encoding.
  Covers: AWS’s P3, G4 series; Azure’s NC-series, ND-series; and GCP’s A2 series.
• High-Performance Compute: For niche use cases like molecular modelling.
  Covers: Azure’s H-series.

‍

Estimating Compute Instance Costs

Each cloud provider offers tools to help estimate costs:

‍

• AWS: Use the AWS Pricing Calculator for precise estimates based on region and usage.
• Azure: Azure Pricing Calculator lets you adjust compute instance options by region, OS, and VM size.
• GCP: The GCP Pricing Calculator helps configure instances for tailored pricing.

‍

Key things to keep in mind:

‍

• Region: Costs can vary by location.
• Discounts: All three clouds offer reserved/committed use discounts. This is where you pay in advance for a fixed time like 1 year or 3 years.
• Spot Instances: All three clouds offer Spot Instances that can save money but may be terminated unexpectedly by the cloud provider.

‍

How Compute Templates Help You?

Managing infrastructure doesn’t have to mean diving deep into the technical weeds. With Wayfinder’s compute templates, you get ready-to-use configurations that make it easier to provision compute instances across multiple clouds. These templates are pre-designed with cost efficiency in mind, so you can focus on building your app, confident that infrastructure costs are under control.

‍

Use Case 1

Imagine being able to spin up a Kubernetes cluster for your app without the back-and-forth of approval tickets. Your platform engineers and finance team have already done the groundwork, so you just choose a compute template from your workspace, with the estimated costs right there to guide you.

‍

• Speed: Focus on coding, not compute setup.
• Simplicity: Choose a template, and Wayfinder takes care of the rest.
• Autonomy: No waiting on approvals - you’ve got the freedom to move ahead.
• Cost Awareness: Make informed decisions without extra complexity.

‍

Use Case 2

Need something more tailored? No problem. You can still configure your own compute instances while using the same guardrails to ensure cost efficiency. The customisation options allow you to adjust settings to fit your exact needs, while Wayfinder’s safeguards help you stay within budget.

‍

• Flexibility: Configure your setup to match your app’s requirements.
• Protection: Guardrails prevent overspending and ensure compliance.
• Control: You have the power to adjust configurations as needed.

‍

What does it look like?

Here’s what to expect when using compute templates:

‍

• Greyed-out templates: These indicate options that are unavailable due to guardrails set by your platform engineers.
• Budget warnings: You’ll receive alerts if your selections risk exceeding the allocated budget.
• Customisation: Templates are flexible, allowing you to adjust settings to fit your exact requirements.

‍

‍

Conclusion

Appvia Wayfinder’s compute templates take the hassle out of cloud infrastructure management. They make provisioning compute instances easy, help you manage costs, and ensure compliance across multiple platforms - all while empowering you to focus on what matters most: building great applications.

‍

If you're a platform engineer and want to learn how compute templates work from your perspective, check out our other post: Advantages of Using Compute Templates for Platform Engineers.