Docs/Vyntech Cloud/Billing & Quotas

Billing & Quotas

Vyntech Cloud is designed with a highly transparent billing engine. This document details how resources are limited (quotas), our transparent per-resource pricing model, and how usage is calculated down to the micro-cent.

Resource Quotas (Limits)

To ensure platform stability and protect users from accidental runway costs, every tenant is subject to resource limits based on their active subscription plan. Before any resource is created (like a new cluster or deploying a workload), the Quota Engine verifies you have sufficient capacity.

ResourceFree Tier (No Subscription)Pro / Enterprise
Clusters1 ClusterDefined by active plan
Workloads (Deployments)3 WorkloadsDefined by active plan
Team Members2 MembersDefined by active plan

Note: If you reach your limit, API and CLI requests will fail with a quota.exceeded error. You must upgrade your plan in the billing dashboard to unlock higher capacity.

How Usage is Calculated (Shared vs BYOC)

Vyntech Cloud supports two modes of deployment, which determine how you are billed:

  • 1.Shared Clusters (Vyntech Pricing): Workloads run on our managed infrastructure. You are billed hourly by Vyntech based strictly on the resources (CPU, Memory, Storage) your workload requests.
  • 2.BYOC / Bring Your Own Cluster (DigitalOcean, AWS, etc): If you connect your own DigitalOcean or AWS account, you pay the cloud provider directly for the underlying servers (e.g., DigitalOcean Droplet pricing). Vyntech does not charge you hourly resource fees for workloads running on BYOC.

Shared Cluster Base Rates

For Shared Clusters, Vyntech charges the following base rates (calculated per hour, billed monthly):

Resource TypeHourly RateEst. Monthly (730h)
CPU (per Core)$0.00548 / hour~$4.00 / month
Memory (per GB)$0.00411 / hour~$3.00 / month
Storage (per GB)$0.00007 / hour~$0.05 / month

Cost Examples (Scenarios)

Because you pay exactly for what you request, here are some common deployment scenarios and their estimated costs:

ScenarioResources RequestedEst. HourlyEst. Monthly (730h)
Microservice API1 vCPU, 512 MB RAM, 10 GB Storage~$0.008 / hr~$6.00 / mo
Web Application1 vCPU, 2 GB RAM, 50 GB Storage~$0.017 / hr~$12.50 / mo
Relational Database2 vCPU, 4 GB RAM, 80 GB Storage~$0.033 / hr~$24.00 / mo
Heavy Backend Worker4 vCPU, 8 GB RAM, 160 GB Storage~$0.066 / hr~$48.00 / mo

The Hourly Worker

Every hour, a background worker scans all active, healthy workloads. It calculates the exact cost incurred during that hour using the following formula (tracked internally in micro-cents to prevent rounding errors):

Cost = (CPU_Request * Replicas * Cluster_CPU_Rate)
     + (Mem_GB * Replicas * Cluster_Mem_Rate)
     + (Storage_GB * Replicas * Cluster_Storage_Rate)

Minimum Charges

  • 1-Hour Minimum: All resources are calculated with a strict 1-hour minimum. You are billed for exactly the resources requested for a full hour, eliminating arbitrary flat-fee minimums (like a rigid $0.01 floor).
  • Termination Rule: When you delete or scale down a workload, a final usage record is generated immediately. Even if the workload ran for just 5 minutes, you are billed for exactly 1 full hour of usage for those resources.

Invoicing & Credits

All hourly usage records are aggregated into a single monthly invoice per tenant.

  • Monthly Generation: On the 1st of every month, our invoice worker aggregates the previous month's usage, applies a flat 10% tax rate, and generates an invoice due in 10 days.
  • Crypto Credits (AzPays): If you use AzPays to top-up your account with cryptocurrency, these credits are deducted in real-time as hourly usage is collected. Your end-of-month invoice will only reflect the remaining unpaid balance (if any).

Real-World Billing Scenarios

To help you understand how our precise pay-per-use billing model applies in practice, we have detailed four common customer profiles. These scenarios demonstrate how workloads are calculated and what the final monthly invoice looks like.

1. The Freelance Developer

Use Case: Hosting a personal portfolio API and occasionally spinning up isolated databases to test weekend side-projects.

Workload Breakdown

  • Portfolio API: 1 vCPU, 512 MB RAM, 10 GB Storage. Runs 24/7 (730 hours).
  • Test Database: 1 vCPU, 1 GB RAM, 25 GB Storage. Spun up on a Saturday, deleted 10 hours later.

End of Month Invoice

ResourceDurationCost
Portfolio API730h~$6.00
Test Database10h~$0.11
Total Invoice$6.11

2. The SaaS Startup

Use Case: Launching an MVP with a highly-available web server, a background worker queue, and a reliable database.

Workload Breakdown

  • Web Server (x2 Replicas): 1 vCPU, 2 GB RAM, 20 GB Storage per replica. Runs 24/7.
  • Background Worker: 1 vCPU, 1 GB RAM, 10 GB Storage. Runs 24/7.
  • Primary Database: 2 vCPU, 4 GB RAM, 100 GB Storage. Runs 24/7.

End of Month Invoice

ResourceDurationCost
Web Server (2 Replicas)730h~$22.04
Background Worker730h~$7.51
Primary Database730h~$25.11
Total Invoice$54.66

3. The Accounting Firm

Use Case: Running internal client portals 24/7, but only spinning up massive payroll calculation nodes at the end of the month to save money.

Workload Breakdown

  • Client Portal: 2 vCPU, 4 GB RAM, 50 GB Storage. Runs 24/7.
  • Payroll Processor: 4 vCPU, 8 GB RAM, 200 GB Storage. Spun up for exactly 5 days (120 hours) during billing week, then terminated.

End of Month Invoice

ResourceDurationCost
Client Portal730h~$22.56
Payroll Processor120h~$8.25
Total Invoice$30.81

4. The Enterprise (BYOC)

Use Case: A massive distributed system using BYOC (Bring Your Own Cluster) to securely manage hundreds of microservices on their own AWS infrastructure.

Workload Breakdown

  • API Nodes: 50 Replicas x (4 vCPU, 8 GB RAM). Runs 24/7 on AWS EKS.
  • ML Workers: 10 Replicas x (8 vCPU, 32 GB RAM). Runs 24/7 on AWS EKS.

End of Month Invoice

ResourceDurationCost
API Nodes (50 Reps)730h$0.00 (Paid to AWS)
ML Workers (10 Reps)730h$0.00 (Paid to AWS)
Vyntech Total Invoice$0.00

We use cookies and similar technologies to measure traffic and improve the site. You can choose which categories to allow. Manage Preferences.