Blog

Notes on Web3 infrastructure

Written by the people running the racks: network policy, oracle economics, data residency and the unglamorous discipline of keeping validators online.

2026.09Operations

Kubernetes network policy: a safe default-deny rollout

Default-deny sounds simple until a liveness probe fails at 3 a.m. How we roll out network policy across validator clusters without taking anything down. The order, the canaries and the rollback plan.

→
2026.09Ecosystem

Chainlink oracle adoption: why platforms standardise instead of building their own

Building an in-house oracle looks cheaper on a spreadsheet. Then you price the security model, the operator set and seven years of uptime. Why platforms keep landing on Chainlink on day one.

→
2026.08Compliance

UK data residency and sovereignty for chain infrastructure

Where your nodes physically sit is becoming a procurement question, not an afterthought. What UK residency actually means for validators and RPC: jurisdiction, access law and audit trails.

→
2026.08Network

What 0.2 ms between nodes buys you

East-west latency rarely makes the marketing page, but it decides how your cluster behaves under load. A walk through our 1.5 Tbps+ fabric and what it means for archive sync and attestations.

→
2026.07Validators

Six years, zero slashing: the operating discipline behind the number

Zero slashing is not luck. It is key management ceremony, change freezes around upgrades, and saying no to things that would look good on a roadmap. A look inside the runbook.

→
2026.07Infrastructure

Bare metal versus cloud for archive nodes: the honest maths

Archive nodes are where hyperscaler billing goes to die. We compare three years of real cost and performance data between our Manchester racks and equivalent cloud instances.

→

Read it first.

Register for VL MKS and get the infrastructure notes, plus launch pricing, before any of it goes public.

Register interest →