Amazon wrote about BSV, and you should pay attention

By Kurt Wuckert Jr.

By Kurt Wuckert Jr. | April 1, 2026

A review and reflection on "How the BSV Association built a million-TPS blockchain node using AWS" by Jordan Kramsky, Senior Solutions Architect for Startups at Amazon Web Services (AWS), published March 31, 2026.

My father used to tell me that you can learn everything you need to know about a man by watching who he works with and how long he sticks around. Not what he says about himself. Not how he markets. Who does he call when there is real work to be done, and does that person pick up the phone?

Yesterday, Amazon (one of the largest companies on earth) published a technical case study on the AWS Web3 Blog about how the BSV Association built a blockchain node capable of sustaining one million transactions per second. Not a press release from BSV people. Not a CoinGeek article. Not my livestream. An Amazon engineer, writing on Amazon's own platform, walking through the architecture of Teranode and the AWS infrastructure that made it possible.

I want you to sit with that for a moment, because I have been writing and talking about this for years, and this is not the kind of validation that comes from a retweet. This is Amazon Web Services telling the world: we helped build this, and we are proud of what it does.

What the article actually says

Jordan Kramsky is a Senior Solutions Architect for Startups at AWS, focused on blockchain and decentralized systems. His article is not a puff piece. It is a detailed technical walkthrough of how the BSV Association's engineering team, led by CTO Siggi Oskarsson, designed and deployed Teranode across six AWS Regions on three continents, sustained one million TPS with zero transaction loss for two weeks straight.

The numbers tell the story.

  1. SV Node The previous BSV reference node, SVNode, peaked at 13,614 TPS
  2. SV Node Optimized Earlier scaling attempts in 2021 reached roughly 50,000 TPS but could not hold the wheels on the digital cart.
  3. Teranode Teranode processes 66.67 times more transactions than SVNode as tested. At 1,000,000 TPS, we are talking about a generational leap.
  4. The Other Guys Visa does 65,000 TPS on Black Friday, and they're the biggest credit card by volume. Solana claims the same throughput as Visa, but over 85% of their TPS volume is reserved for finding consensus. Teranode is orders of magnitude more scalable today.

Kramsky details the specific AWS services involved: storage-optimized EC2 instances with NVMe SSDs for UTXO management, EKS for container orchestration, FSx for Lustre for high-performance shared data across microservices, S3 for durable long-term storage, VPC peering across regions, and managed observability through Prometheus and Grafana. The BSVA team even built a custom Kubernetes operator to standardize deployments and manage scaling. They used Aerospike on NVMe volumes as a memory buffer for low-latency UTXO operations during block creation and propagation.

The architectural breakthrough is what the article calls "subtree broadcasting." Traditional blockchains use episodic batch validation: transactions pile up in a mempool and get processed all at once when a block is mined. Teranode does something fundamentally different. It continuously broadcasts Merkle subtrees every second, pre-validating transactions and assembling block components in real time across the network. Block assembly transforms from a punctuated event into a fluid, continuous operation. Kramsky writes that these subtrees contained up to a million transaction IDs with complete Merkle path connectivity, all verifiable by network nodes.

This is how you go from "blockchain does not scale" to "blockchain is infrastructure." You rethink the problem from first principles and build like the future depends on it!

Why this matters more than the tech

I have written before that the path to a million TPS was not a fantasy, but a matter of engineering will and institutional focus. When I wrote about that in 2024, people in the broader crypto space laughed. They said BSV was dead. They said nobody cared. They said a million TPS was a vaporware promise from scammers.

The people who said that were not building anything. They were trading tokens and arguing about memecoins.

And that brings me to the real point.

You become who you work with

The BSV Association (a Swiss non-profit) made a decision years ago to partner with AWS rather than trying to do everything themselves on bare metal in a closet somewhere. That decision was strategic and, honestly, philosophical. It said: we are building something that needs to work at global scale, and we are going to work with the company that operates the world's largest cloud infrastructure to do it.

AWS, for their part, did not just hand over servers and walk away. Kramsky's article describes AWS Support helping the Teranode team fine-tune operating system and kernel settings to optimize network and storage performance. At the AWS Zurich Summit in 2024, Oskarsson praised the relationship directly: AWS helped them get this done quickly, and the partnership made a five-year R&D journey achievable.

This is what real collaboration looks like. Not a logo on a slide deck. Not a "strategic partnership" announcement that means nothing. Engineers from two organizations, working together on hard problems, iterating through failures, and arriving at a result that neither could have achieved alone. The BSVA brought the protocol expertise and the novel subtree architecture. AWS brought the global infrastructure, the managed services, and the support engineering to make it performant at scale.

I have seen this pattern play out in every domain I have been involved in. In my own experience with GorillaPool, the first team to mine a GorillaNode block on mainnet in December 2025, success came from staying close to the people who were actually doing the work and keeping the lines of communication short and honest.

What this means for Bitcoin going forward

The crypto industry has spent a decade arguing about whether blockchain can scale. BTC chose to abandon on-chain scaling in favor of Layer 2 solutions and a "store of value" narrative. Ethereum went to proof of stake and sharding. Solana optimizes for speed but accepts trade-offs on decentralization. None of them have demonstrated one million sustained TPS with zero loss on a proof-of-work UTXO chain that follows the original Bitcoin protocol rules.

Teranode did. And Amazon wrote about it on their own blog.

This has implications for AI agent infrastructure, for enterprise data systems, for micropayments, for digital identity, for supply chain management, and for every use case that requires a high-throughput, low-cost, regulation-ready ledger. As I have argued for years, the technology was never the bottleneck. The bottleneck was always the willingness to take the work seriously and partner with serious people.

The BSV Association took the work seriously. AWS took the work seriously. And, without patting myself on the back a little too much, GorillaPool also took the work seriously as the first, independent tester of the Teranode software on a public blockchain. The result is open source, publicly available on GitHub, and documented by one of the world's largest technology companies.

A final thought

I have been covering Bitcoin for years. I have watched projects rise and fall based on whether they could sustain focus long enough to deliver real results. The projects that succeed share something in common: they find the right partners, they stay disciplined, and they do not get distracted by the noise.

The BSVA-AWS relationship is a textbook case. Five years of iterative development. A team that was willing to throw out assumptions that were not working and start fresh with new architecture. A cloud partner that provided not just infrastructure but active engineering support. And at the end of it, a published result that speaks for itself.

No hype cycles. No token launches. No influencer campaigns. Just engineering, collaboration, and results.

Be good to each other.

Footnotes

¹ Jordan Kramsky, "How the BSV Association built a million-TPS blockchain node using AWS," AWS Web3 Blog, March 31, 2026.

² SVNode peak throughput of 13,614 TPS and Teranode's 66.67x improvement as reported in the AWS case study.

³ The BSV Association is a Swiss non-profit organization that coordinates technical standards, developer education, and commercial adoption for the BSV blockchain.

⁴ Teranode source code and tooling are publicly available at the BSV Blockchain GitHub repository.

⁵ GorillaPool mined the first Teranode block on BSV mainnet on December 15, 2025. See "GorillaPool Mines First Teranode Block: A Massive Milestone," CoinGeek, December 18, 2025.

⁶ Siggi Oskarsson's remarks at the AWS Zurich 2024 Summit as quoted in the AWS case study.

Entity references

Related reading