Google Cloud sunsets Blockchain Node Engine on December 15, hands customers to Quicknode

Google Cloud will shut down its Blockchain Node Engine (BNE) and the companion Blockchain RPC service on December 15, 2026, Crypto Briefing reports. Google has named Quicknode the recommended replacement and published the official migration path in its own docs at cloud.google.com/blockchain-node-engine/docs/migrate-to-quicknode. All remaining nodes and endpoints will be automatically deleted on the shutdown date.

The two shutdown phases

• June 15, 2026 (already in effect): Google stopped accepting new BNE node creation requests and halted provisioning of new Blockchain RPC endpoints. Existing resources continued running and received critical updates only.
• December 15, 2026 (full termination): Every remaining node and RPC endpoint is deleted. Any application still pointing at a BNE URL after that moment starts returning errors.

What BNE covered

BNE provided managed full and archive nodes for:

• Ethereum — mainnet and testnets.
• Polygon — mainnet.
• Solana — mainnet.

The hosted Blockchain RPC service — a lighter, pay-per-request endpoint — was scoped more narrowly to Ethereum mainnet and the Holesky testnet.

Archive nodes — those tracking the full historical state rather than only the current tip — carry the slowest migration path. Replicating a complete archive is bandwidth-heavy, and the authoritative state has to be re-verified from genesis or sourced from an existing archive peer; it is not something that can be left until the week before shutdown.

The Quicknode arrangement

The migration can be run through two routes:

1. Google Cloud Marketplace — carries Google-side incentives on top of Quicknode's pricing.
2. Direct Quicknode onboarding — standard sign-up on the provider's platform.

Quicknode is one of several providers with equivalent coverage; Chainstack has also published a migration guide for former BNE customers, and Alchemy, Infura (ConsenSys) and Blockdaemon serve the same workloads. The "recommended" tag is a distribution deal, not a protocol constraint — customers are free to migrate anywhere.

Context

The shutdown is the second major cloud-side blockchain exit in a short window. Managed-node services are expensive to run at the SLA institutional customers demand, and the margins have never been as attractive as the broader cloud business. Google's post-mortem framing — "handing customers to Quicknode" — is explicit about the reason: the specialist provider is a better fit for the workload than a hyperscaler.

For teams running production infrastructure, this is a hard migration deadline. There are no indications Google will extend it.

Action checklist

1. Inventory every service that references a BNE URL — RPC endpoints, archive readers, indexers, custom validators. Mixed configs where a few microservices still point at BNE while others moved years ago are the common failure mode.
2. Migrate archive-node workloads first. Historical replication is the long pole.
3. Decide the migration target now, not in December. Quicknode is the Google-blessed path; Chainstack, Alchemy, Infura and Blockdaemon are equivalent-tier alternatives. Price sensitivity, geographic latency and the mix of chains you serve are the three variables that pick between them.
4. Validate the new endpoints under production load — RPC method coverage, eth_subscribe behavior, archive query depth, rate limits — before cutting over.
5. Set a firm cutover date well before December 15. Treat the Google deletion as a hard failure, not a graceful degradation.