Validators
General considerations
Please check out the security considerations before using Nethermind as a validator.
For Ethereum validators, we highly recommend checking out Staking with Ethereum and Validator checklist.
Hardware configurations
The following hardware configurations for Ethereum Mainnet validators have been battle-tested by us and our users. We have observed excellent validator performance and stability with these configurations.
Before setting up your infrastructure, check out Nethermind hardware requirements.
On-premises
A single validator on Intel NUC 11:
- CPU: Intel Core i7-1165G7
- Memory: Crucial 32GB DDR4-3200 SODIMM
- Storage: Samsung 980 PRO PCIe NVMe SSD 2TB
- Internet speed: 620 Mbps download, 160 Mbps upload
AWS
Multiple validators on the following EC2 instances:
- m6i.2xlarge: 8 vCPU, 32 GiB memory
- m7g.2xlarge: 8 vCPU, 32 GiB memory
These configurations have proven to work well for 1000-1500 validators and haven't been tested for more validators. Also, the validator clients have been separated from the consensus and execution clients and running on t4g.small instances.
Azure
Multiple validators on the following VM instances:
- Standard_D8_v5: 8 vCPU, 32 GiB memory
- Standard_D8ps_v5: 8 vCPU, 32 GiB memory
These configurations have proven to work well for 1000-1500 validators and haven't been tested for more validators. Also, the validator clients have been separated from the consensus and execution clients and running on Standard_D2pls_v5 instances.
GCP
Multiple validators on the c2d-highmem-4 instance: 4 vCPU, 32 GB memory
These configurations have proven to work well for 1000-1500 validators and haven't been tested for more validators. Also, the validator clients have been separated from the consensus and execution clients and running on e2-small instances.
Gnosis validators
To set up a Gnosis Chain validator, you can either do that manually or use one of the available one-click tools.