Skip to main content
The Credible Layer operates as a sidecar alongside the network sequencer. This page outlines the system requirements for running the sidecar.

Production Requirements

For production usage on a real network, run the sidecar with at least the following specifications:
ComponentRequirement
CPU32 physical cores, AMD Zen 3 performance equivalent or higher
RAM128 GB recommended
Storage512 GB+ recommended
Network10 Gbps minimum between sequencer and sidecar

CPU

The sidecar requires significant compute for assertion execution. Use 32 physical cores with AMD Zen 3 performance equivalent or higher.

Memory

Allocate 128 GB RAM minimum. RAM allocation should be enough to store the entire chain state (not full blocks, just state) in memory.

Storage

Use 512 GB+ of fast local (not networked) PCIe NVMe SSDs to keep I/O latency low. Budget for sustained IOPS and 2x storage capacity to maintain fast SSD reads.

Networking

Keep round-trip time (RTT) to the sequencer sub-millisecond by colocating in the same availability zone or cluster. Co-scheduling on the same Kubernetes node/pod is ideal. Key requirements:
  • 10 Gbps minimum bandwidth between sequencer and sidecar
  • VPC-peer the sequencer and sidecar; avoid routing over the public internet
  • Stable, low-jitter egress to your DA RPC (HTTP) and indexer RPC (WebSocket) endpoints
  • Prefer private endpoints or allowlisted static egress IPs

Local Development

For local testing, unless you are benchmarking performance, the sidecar is lightweight and has no hard hardware requirements.