1. What is NodeCore?#

NodeCore is an open-source, self-hosted RPC infrastructure stack that lets teams run and control their own blockchain connectivity layer.

At its core, NodeCore acts as a high-performance load balancer and routing engine, distributing traffic across multiple nodes and providers to optimize latency, reliability, and cost — without relying on managed SaaS RPC services.

It is designed for teams that require infrastructure sovereignty, performance tuning, and vendor independence while maintaining production-grade resilience.

Why teams deploy NodeCore#

Organizations adopt NodeCore to:

• Eliminate dependency on single RPC providers
• Control routing logic and performance policies internally
• Reduce infrastructure costs at scale
• Run hybrid setups combining self-hosted and third-party nodes
• Maintain uptime during upstream outages

Design Goals#

NodeCore is built around four core infrastructure principles:

1. Cost Efficiency

Smart caching combined with price-aware routing minimizes redundant upstream calls and optimizes provider spend without compromising performance.

2. Performance-Aware Routing

NodeCore continuously evaluates provider latency, error rates, and throughput to select the most suitable upstream for every request. Underperforming providers are automatically deprioritized.

3. Resilient Error Handling

Automated retries, hedged parallel requests, and circuit-breaker logic protect applications from upstream instability and long-tail latency spikes.

4. Operational Simplicity

NodeCore centralizes routing, usage analytics, and provider monitoring into a unified control layer. It simplifies multi-provider infrastructure management.

Who is NodeCore for?#

Best fit for:

• Web3 infrastructure and DevOps teams running their own nodes
• dApps with backend-originated traffic from one or few regions
• Enterprises requiring on-prem or VPC-contained RPC traffic
• Teams combining multiple RPC vendors for redundancy and cost control

Less ideal for:

• Applications with globally distributed, browser-originated traffic
• Frontend-heavy dApps requiring edge routing worldwide
• Teams seeking fully managed, no-ops RPC infrastructure

In these cases, a managed global routing solution such as NodeCloud may be more appropriate.

Key Features#

1. Intelligent Cost-Aware Routing

Routes requests using real-time latency, error-rate, and pricing metrics to balance performance and cost.

2. Multi-Chain / Multi-Protocol Support

Compatible with Ethereum-like chains, Solana, Cosmos, Bitcoin, TON, Polkadot, Near, Starknet, and more across JSON-RPC, WebSocket, gRPC, and REST.

3. Aggressive, Correctness-Aware Caching

Supports memory, Redis, and Postgres caching layers to reduce redundant traffic and upstream load.

4. Real-Time Metrics & Observability

Exports Prometheus metrics and OpenTelemetry traces out-of-the-box for deep performance monitoring.

5. Streaming-First Response Handling

Streams large payloads by default to minimize memory footprint and improve throughput.

6. Resilient Error Handling

Retries, hedging, and circuit-breaker protections ensure consistent uptime.

7. NodeCloud Bridge (Optional)

Sync API keys, usage budgets, and analytics with NodeCloud for hybrid deployments.

System Architecture#

NodeCore sits between client applications and upstream RPC providers:

Client → NodeCore Load Balancer → Cache Layer → Providers → Metrics & Tracing

Reliability Design Principles#

NodeCore is built around production resilience:

• No single-provider dependency (N+1 routing)
• Hedged reads to reduce long-tail latency
• Circuit breakers eject failing nodes
• Checksum-verified read caching
• Automatic rate-limit discovery

Deployment Scenarios#

Common production setups include:

• Routing across Infura + Alchemy + self-hosted nodes
• Backend traffic balancing from a single region
• Hybrid NodeCore + NodeCloud failover
• Private RPC infrastructure for enterprise workloads