Running a Plasma node requires thoughtful hardware selection to ensure performance, reliability, and cost-efficiency. This guide outlines hardware recommendations for various deployment scenarios, focusing on non-validator nodes while consensus participation remains unavailable during testnet.
Contact required: Please submit your details here before deploying nodes.

​
Development vs Production Environments

Hardware needs vary between development and production. Choose the appropriate tier to balance cost, performance, and operational risk.

​
Development & Testing

Use lower-cost hardware for experimentation, prototyping, or local development. Minimum Specifications:
  • 2 CPU cores (any modern architecture).
  • 4 GB RAM.
  • 100 GB SSD storage.
  • Standard internet connection (10+ Mbps).
Suitable Cloud Instances:

Amazon Web Services

t3.small t3.medium

Google Cloud Products

e2-small e2-medium

Microsoft Azure

B2s B2ms
Cost Considerations:
Development nodes can use burstable instances for cost savings. Spot instances can be suitable for non-critical testing. Local development on laptop/desktop machines often sufficient.

​
Production Deployments

For applications requiring reliability, low latency, and uptime guarantees. Recommended Specifications:
  • 4+ CPU cores (high clock speed preferred).
  • 8+ GB RAM.
  • 500+ GB NVMe SSD storage.
  • Low-latency internet connection (100+ Mbps).
  • Redundant network connectivity preferred.

​
Cloud Provider Recommendations

Below are provider-specific instance suggestions for development and production use cases:

​
Amazon Web Services (AWS)

AWS provides extensive instance options with global availability and mature blockchain infrastructure support.Development instances:
  • t3.small: 2 vCPUs, 2 GB RAM, variable performance.
  • t3.medium: 2 vCPUs, 4 GB RAM, burstable CPU credits.
Production instances:
  • c6a.large: 2 vCPUs, 4 GB RAM, high CPU performance.
  • c6a.xlarge: 4 vCPUs, 8 GB RAM, consistent high performance.
  • m6i.xlarge: 4 vCPUs, 16 GB RAM, balanced compute and memory.
Storage recommendations:
  • gp3 volumes for cost-effective performance.
  • Provisioned IOPS for high-traffic RPC endpoints.
  • Regular snapshots for data protection.

​
Google Cloud Platform (GCP)

GCP provides competitive pricing and strong performance, particularly for compute-intensive workloads.Development instances:
  • e2-small: 2 vCPUs, 2 GB RAM, cost-optimised.
  • e2-medium: 2 vCPUs, 4 GB RAM, balanced performance.
Production instances:
  • c2-standard-4: 4 vCPUs, 16 GB RAM, high-performance CPU.
  • n2-standard-4: 4 vCPUs, 16 GB RAM, balanced workloads.
  • c2-standard-8: 8 vCPUs, 32 GB RAM, high-traffic scenarios.
Storage recommendations:
  • SSD persistent disks for all deployments.
  • Higher IOPS allocation for production nodes.
  • Automatic snapshots for backup and recovery.

​
Microsoft Azure

Azure offers competitive pricing and integrated services particularly suitable for enterprise deployments.Development instances:
  • B2s: 2 vCPUs, 4 GB RAM, burstable performance.
  • B2ms: 2 vCPUs, 8 GB RAM, higher baseline performance.
Production instances:
  • F4s_v2: 4 vCPUs, 8 GB RAM, compute-optimised.
  • D4s_v5: 4 vCPUs, 16 GB RAM, general purpose.
  • F8s_v2: 8 vCPUs, 16 GB RAM, high-performance option.
Storage recommendations:
  • Premium SSD for production workloads.
  • Standard SSD acceptable for development.
  • Azure Backup for automated data protection.

​
Getting Started

To begin running a Plasma non-validator node with appropriate hardware:
  1. Assess your requirements: Determine whether you need development or production-grade infrastructure.
  2. Submit your details: Contact our team to receive deployment information.
  3. Select cloud provider: Choose based on your geographic needs, existing relationships, and pricing preferences.
  4. Configure monitoring: Set up essential monitoring before deployment to track performance from the start.
  5. Deploy incrementally: Start with minimum viable specifications and scale based on actual usage patterns.
  6. Plan for growth: Design your infrastructure to accommodate future scaling needs without major architectural changes.
For detailed deployment instructions and configuration guidance, proceed to the Overview and Non-Validator Node Setup sections.