[ANN] 10 years of Bitcoin history, replayed in under 30 minutes

[Witos89]

Today we replayed the entire 10 years of Bitcoin’s transaction history on the Radix ledger, with full transaction and signature validation, on a network of over 1,000 Nodes distributed evenly throughout the world.

For the first time since the creation of public, trustless networks, we have demonstrated a technology that can truly support even the world’s most demanding transactional applications using.

Comparison between platforms in terms of TPS - https://www.radixdlt.com/wp-content/uploads/2019/06/design-7.png


The Result?

The Radix ledger peaked at over 1 million transactions per second, a record for decentralized ledgers. It is even faster than other centralized and established payment processors like Alipay, WeChat and PayPal to name a few. The whole dataset was processed in under 30 minutes.

We were able to achieve this due to Radix’s unique design. Instead of using Blockchains or DAGs, we started from scratch, building a new database and consensus mechanism called Tempo that can scale to support 7.5 billion people and 500 billion devices, simultaneously.

After 7 years of research and development, many dead ends and sleepless nights, today represents a significant milestone in the Radix journey. We can proudly say we’ve built a full state-sharded decentralized ledger that is massively scalable without compromising on security.

To validate our engineering we have decided to run a series of public network throughput tests in an open, transparent manner, that anyone can verify and see happen live.

We decided to use Bitcoin as our data source because it has processed a large number of transactions over the last decade (over 400 million transactions and 460 million addresses) and is an open, fair and transparent data set.

As a comparison, you could run applications like VISA, PayPal, WhatsApp, WeChat and more, simultaneously on the Radix ledger, without breaking a sweat.

The first public test happened today 14.30 London time.

The time and date of our next live test can be seen on our explorer - https://test.radixdlt.com/. Make sure you also have your bitcoin addresses ready, as when the test is running you will be able to add in your address and see your transaction history replayed in real time.

These tests will run approximately every week from now!  Please sign up to the Radix newsletter and get notified before the next test starts: https://radixdlt.typeform.com/to/x3yotG

Interested in how we did it?

The following technical blog posts explain our methodology and the step-by-step chronology of how we achieved this throughput.

Part 1: A primer on the scalability test and Radix (non-technical) - https://www.radixdlt.com/post/test-method-part1/
Part 2: How we actually built the scalability test (technical) - https://www.radixdlt.com/post/test-method-part2/
Lastly, if you have any questions, or just want to follow us as we go towards the mainnet, please join the conversation with our community on Telegram (https://t.me/radix_dlt) or use the #1MTPS on Twitter (https://twitter.com/radixdlt) to share your opinions.

[1713969948]

1713969948

[1713969948]

1713969948

[wingspan]

Back in 2015 I offered to run a speed test (Tx/Sec) "bakeoff" to see which altcoin was fastest.  My favorite project (eMunie at the time - now called "RadixDLT") stood unopposed so no bakeoff happened - perhaps due to my approach.   https://bitcointalk.org/index.php?topic=1203228.0
After 4+ years, I see eMunie/RadixDLT did a speed test similar to what I was thinking - but larger.  It was not a perfect test of what their code would be able to do in production, but it was pretty amazing none-the-less.
Does anyone know of another altcoin that can come close in speed, security, and decentralization in a speed test?   Would they be willing to participate in a public bakeoff of any sort?  Let me know if I can help.
I also appreciate that speed is not everything.  But it is one requirement of many before a cryptocurrency can become globally adopted.
-cheers,
- Wingspan.

[joniboini]

Does anyone know of another altcoin that can come close in speed, security, and decentralization in a speed test?   Would they be willing to participate in a public bakeoff of any sort? 

There are a lot of scalability focused project in the market right now, but not sure about their speed. I think Elrond Network, Matic, Zilliqa, is a few of them. Not to mention EOS and so on.

Zilliqa is probably the most mature in terms of development, their mainnet is alive, while EOS, as you might know, is too centralized.

[wingspan]

...
There are a lot of scalability focused project in the market right now, but not sure about their speed. I think Elrond Network, Matic, Zilliqa, is a few of them. Not to mention EOS and so on.

Zilliqa is probably the most mature in terms of development, their mainnet is alive, while EOS, as you might know, is too centralized.

Thank you, I'll dig into Elrond, Matic, and Zilliqa for max speed info if any - or if a bakeoff would interest those projects.
- Wingspan

[13Darko]

Does anyone know of another altcoin that can come close in speed, security, and decentralization in a speed test?   Would they be willing to participate in a public bakeoff of any sort?  

There are a lot of scalability focused project in the market right now, but not sure about their speed. I think Elrond Network, Matic, Zilliqa, is a few of them. Not to mention EOS and so on.

Zilliqa is probably the most mature in terms of development, their mainnet is alive, while EOS, as you might know, is too centralized.

Current speed:

Elrond - 36,000 TPS on 16 shards - comment from admin in their official Telegram group dated by 4th of July 2019
Matic   - 6,000 - 10,000 TPS in closed environment, and a single sidechain is theoretically capable of up to 65K TPS - comment from admin in their official Telegram group dated by 1st of July 2019
Zilliqa  - 2,828 TPS on testnets, with shards, 2400 nodes - comment from admin in their official Telegram group dated by 4th of July 2019
EOS    -  centralized, but 9,179 TPS according to this article dated by 05.12.2018: https://medium.com/@cryptolions/new-maximum-eosio-tps-demonstrated-in-jungle-testnet-9179-15a485f2e79

Radix - 1,417,786 TPS (https://twitter.com/radixdlt/status/1146423706958344194) on ~1200 nodes spread across the world.

[leea-1334]

Current speed:

Elrond - 36,000 TPS on 16 shards - comment from admin in their official Telegram group dated by 4th of July 2019
Matic   - 6,000 - 10,000 TPS in closed environment, and a single sidechain is theoretically capable of up to 65K TPS - comment from admin in their official Telegram group dated by 1st of July 2019
Zilliqa  - 2,828 TPS on testnets, with shards, 2400 nodes - comment from admin in their official Telegram group dated by 4th of July 2019
EOS    -  centralized, but 9,179 TPS according to this article dated by 05.12.2018: https://medium.com/@cryptolions/new-maximum-eosio-tps-demonstrated-in-jungle-testnet-9179-15a485f2e79

Radix - 1,417,786 TPS (https://twitter.com/radixdlt/status/1146423706958344194) on ~1200 nodes spread across the world.

Good statistics collected,,, but I am now thinking, if only there was a rating site that tested out all of these networks and blockchains to see if TPS was really true. I agree, speed is not everything but it is something. It will not be a problem for a long time anymore after Bitcoin gets the Lightning Network to be adopted too.

Should mention Waves as well right?

[13Darko]

Current speed:

Elrond - 36,000 TPS on 16 shards - comment from admin in their official Telegram group dated by 4th of July 2019
Matic   - 6,000 - 10,000 TPS in closed environment, and a single sidechain is theoretically capable of up to 65K TPS - comment from admin in their official Telegram group dated by 1st of July 2019
Zilliqa  - 2,828 TPS on testnets, with shards, 2400 nodes - comment from admin in their official Telegram group dated by 4th of July 2019
EOS    -  centralized, but 9,179 TPS according to this article dated by 05.12.2018: https://medium.com/@cryptolions/new-maximum-eosio-tps-demonstrated-in-jungle-testnet-9179-15a485f2e79

Radix - 1,417,786 TPS (https://twitter.com/radixdlt/status/1146423706958344194) on ~1200 nodes spread across the world.

Good statistics collected,,, but I am now thinking, if only there was a rating site that tested out all of these networks and blockchains to see if TPS was really true. I agree, speed is not everything but it is something. It will not be a problem for a long time anymore after Bitcoin gets the Lightning Network to be adopted too.

Should mention Waves as well right?

Exactly. These tests have to be:
1) verifiable
2) non-laboratory (nodes have to be located on different continents)
3) with full signature verification (not like other projects usually do: Hashgraph didn't verify signatures,  for instance - that's written in their whitepaper)
4) transaction size should be the same as it will be on main net - definitely more than >100 bytes.

Radix achieves all:
1) this test is verifiable (you can repeat it: https://www.radixdlt.com/post/test-method-part2/), it was replayed many times already and every time you could see ~1M tps. Also these tests are done using Bitcoin transactions database, so everyone can check any Bitcoin address during a test.
2) it was done on >1000 nodes in 17 cities around the world - decentralized.
3) it did full signature verification
4) average transactions size = 1 KByte

Speed/scalability is not just "something", it's a real issue with current blockchains/DAGs. You can't and you won't have mass adoption until this problem is solved. You need billions of people/devices to work simultaneously. If you want WeChat, WhatsApp, payment rails (like VISA, Alipay etc.), IoT devices to work all on top of the same ledger, it needs to be really scalable, without bottlenecks, permissionless and secure. Something which solves the Scalability trilemma. That is Radix.

Lignting Network will help Bitcoin a little, but its blockchain will still be a bottleneck. Think about it youself - when millions of people will open/close channels in LN, everytime there will be a record/transaction to the blockchain, which can't settle more than 7 txs per second. So it will work only until it hits a limit.
Ideally transactions need to be done on-chain. You simply don't need 2nd layer solutions if you have scalable tech.

Waves doesn't solve a scalability problem.