Blog

Latest news and insights about Web3 infrastructure, liquid staking, and onchain finance.

Kubernetes network policy: a safe default-deny rollout: A production rollout guide for Kubernetes NetworkPolicy. Confirm your CNI enforces it, start from observed flows, deny ingress before egress, and keep a rollback.
Chainlink oracle adoption: why platforms standardise instead of building their own: Wyoming and Nethermind both chose Chainlink over their own cross-chain infrastructure in August 2026. What standardising on an oracle network actually asks of the node operators underneath it.
UK data residency and sovereignty for chain infrastructure: UK data residency usually stops at which region a workload runs in. For chain infrastructure, where the assets at stake are keys rather than data, the real questions are legal jurisdiction and physical auditability.
European cloud providers: what to compare before you choose: A requirements-first comparison of European cloud providers for teams running RPC nodes, validators, indexers, oracle services and other stateful infrastructure.
What a staking ETF demands of the infrastructure underneath: Morgan Stanley listed ETH and SOL staking products on NYSE Arca at a 0.14% expense ratio. The filing terms describe, in numbers, what an issuer will accept from a validator operator.
What is an RPC node, and what it takes to run one on a busy chain: An RPC node is a server that lets apps and wallets read a blockchain and broadcast transactions. What one does, how it differs from a validator node, and what it actually requires once the chain it serves gets busy.
Taking a Solana validator through a client upgrade: what actually breaks: Solana is running at record load, which raises the cost of a validator falling over during a routine client upgrade. What actually breaks, and how to stage a reversible rollout so record throughput never turns into missed slots.
What application revenue tells you about which chains to run infrastructure on: Chain application revenue is the hardest metric to fake. Why installs and TVL mislead, and what revenue-led chains demand of validator and RPC infrastructure.
Self-hosted RPC vs managed RPC: cost and control trade-offs: When should a protocol team run its own RPC endpoint and when should it buy managed? A breakdown of the real cost and control trade-offs, from sync maintenance and egress to the reliability gap users actually feel.
Where RWA growth is happening onchain and what it demands of infrastructure: Real-world-asset activity is concentrating on a few networks. Where onchain RWA growth is showing up, and what institutional settlement demands from the infrastructure underneath.
Archive node hosting cost: cloud vs dedicated infrastructure in 2026: An Ethereum archive node grows continuously, and on shared-tenant cloud the storage bill compounds with it. Why archive node hosting cost looks fine in month one and breaks by month twelve, on cloud versus dedicated infrastructure.
DTCC on Stellar: what tokenised TradFi means for chain infrastructure demand: DTCC is putting tokenised stocks, ETFs, and Treasuries on Stellar. What TradFi settling on a public chain means for validator selection, RPC procurement, and institutional infrastructure due diligence.
DVT stopped being thought leadership in March 2026: The Ethereum Foundation staked 72,000 ETH using DVT-Lite. AWS Marketplace lists DVT-as-a-Service. Vitalik publicly pushed DVT-Lite as the default. Threshold signing is no longer a position. What institutional allocators should ask now.
Building in-house staking and oracle infrastructure in 2026: What it actually takes to build and operate Ethereum validator and oracle node infrastructure in-house in 2026. Hardware, key management, DVT, monitoring, staffing, and the operational load most teams underestimate.
How to evaluate managed Ethereum staking and oracle operators in 2026: What institutional buyers should look for when evaluating managed Ethereum staking operators and oracle infrastructure providers: infrastructure model, jurisdiction, slashing history, DVT readiness, and protocol coverage.
Why in-house institutional staking is harder than it looks: The structural reasons institutional Ethereum staking is difficult to operate in-house: client diversity requirements, hard upgrade deadlines, DVT coordination, staffing gaps, and the long-tail incidents that accumulate before anyone notices.
RPC endpoint hosting cost: AWS vs dedicated infrastructure (2026): A single RPC endpoint workload costs $1,112 per month on AWS before egress, and $219 on the cheapest whole server. Side-by-side pricing across AWS, Hetzner, Latitude and Servers.com, with the egress and state-bloat context behind the gap.
Why Visa running a validator is the more interesting validator story: Two validator stories landed in April 2026. The more structurally interesting one is the shift from institutions delegating staking to institutions running validator infrastructure themselves. What that shape requires operationally, and where the hardware question lands.
What goes into running a Chainlink Data Streams node: Chainlink Data Streams now carries U.S. stocks and ETFs on a 24/5 basis. Here is what running a Data Streams node requires operationally: hardware, network latency, uptime, and failover.
ETH validator hosting cost: AWS vs dedicated infrastructure (2026): One ETH validator workload costs $556 per month on AWS and $219 on the cheapest whole server. Side-by-side pricing for AWS, Hetzner, Latitude and Servers.com, and why the pricing model matters more than the sticker price.
What Lido's Identified DVT Cluster vote means for independent operators: Lido's Identified DVT Cluster vote closed April 13. Four independent operators, distributed key generation, 3-of-4 threshold signing. What it means for operators running the hardware behind the yield.
Infrastructure considerations for MiCA compliance: what changes after July 1, 2026: The EU MiCA grandfathering period ends July 1, 2026. Infrastructure checklist for CASPs: jurisdiction documentation, operational resilience, key management, physical auditability, and what a compliance review will ask for.
Bare-metal Kubernetes for Web3: we ran it before enterprise caught up: Why Web3 operators run bare-metal Kubernetes for validators, Chainlink nodes, and RPC endpoints, and how the cost compares to AWS and GCP today.
What is distributed validator technology (DVT)?: DVT splits Ethereum validator control across multiple independent operators using threshold signing. No single key, no slashing risk, no single point of failure. How SSV Network, Obol, and Lido DVT works, and what to ask before delegating institutional ETH.
Why We Chose Bare Metal Kubernetes for Web3 Infrastructure: How LinkPool runs high-performance Kubernetes clusters on bare metal servers across three availability zones in Manchester, and why cloud was never an option for mission-critical Web3 workloads.