Sperax introduces our inaugural virtual event series in the last week of every month via all of our global channels — — Chinese, Korean and English. In the near future, Sperax team will go over the company history, stable coin adoption, China’s perspective on consensus distribution, etc.
For our very first panel, Sperax strategy advisor Helen Ruan hosted the event on May 28, 2020 together with Sperax researcher Joe Yu from UCLA and amazing attendees from FBG Capital, Blockchain Capital and more, who raised in-depth questions. From a technical standpoint, Joe covered Sperax blockchain consensus protocol and how it self-differentiates from some well-known competitors.
For those of you who did not make it, we got you covered. Or you can watch the recording here:
Our design principle:
We believe the best blockchain applications should not require users to interact with the blockchain itself.
We aim to achieve security and performance at the same time. Our consensus protocol is the most efficient among BFT-based consensus.
We want to connect a crypto-based decentralized economy with real world finance. Our solution includes a native stablecoin to facilitate the adoption of non-crypto users.
What’s Unique about Sperax BDLS Consensus Protocol?
The Model of the Internet is an open network which is better modelled by a partially synchronous network model. It alternates between a synchronous network and an asynchronous one with an unknown transition time called Global Stabilization Time (GST). We analysed many widely deployed BFT protocols for blockchains and to our best knowledge, all of them could go to deadlock before GST and then remain deadlock (even after synchrony).
Sperax BDLS protocol is designed based on a partially synchronous network model (more realistic). So our protocol can provide a more secure method of consensus. Our protocol doesn’t assume reliable communication channels for each node pair in the network. Thus, we take the possibility of DoS attack in a partially synchronous network into account. Since our protocol is a blockchain version of the famous DLS protocol in the traditional cryptography literature, our protocol is thus named the BDLS protocol.
BDLS has 2 major advantages:
Security — It is mathematically proven that BDLS protocol could achieve safety and liveness property in partially synchronous networks. It is known that safety and liveness are the two basic security requirements for a Byzantine Fault Tolerance (BFT) protocol.
Performance — Faster than all existing BFT protocols for partial synchronous networks
(If you’d like to learn more, feel free to check out the BDLS protocol paper.)
Comparison No.1: Tendermint BFT
According to recent statistics, Tendermint BFT is used by around 40% of PoS blockchain systems, including many interesting projects such as Cosmos blockchain, Oasis Labs, and Binance DEX.
Security vulnerabilities within Tendermint BFT: not secure and inefficient in communications, hence lags in performance.
Tendermint’s protocol consists of 3 steps at each block height: propose, prevote and precommit, where the initiator of each round is determined by a public function.
If a malicious node becomes the initiator, it can successfully carry out an attack so that each node in the blockchain network will lock a different value before the GST is reached. These locked values cannot be removed even after GST. Thus, the whole system goes into a deadlock. And no new block will be produced.
Comparison No.2: Casper FFG
Casper FFG works on a “epoch tree.” An epoch is 50 consecutive blocks as a whole. These epochs are organized in a tree structure, and so we get the name “epoch tree.” A block in the Casper FFG blockchain is considered finalized only when the epoch containing that particular block is validated and finalized.
Security vulnerabilities within Casper FFG: relatively fast but not secure.
Without further restrictions on the block proposal mechanism, Casper FFG could reach a deadlock. During the unreliable network period, an underlying block may reach some validators though not other validators. Now assume that t honest validators receive a descent block B and vote for it while other (t+1) honest validators receive a descent block B’ and vote for B’ instead. Assume that the remaining t validators are all malicious and would not vote at all. Then neither B or B’ can get sufficient votes to be finalized. Thus, we go into a deadlock and this deadlock could not be removed.
Ethereum 2.0 FORKED less than 2 weeks ago. Given that forking should be absolutely avoided in the blockchain world, the Ethereum team agreed that it was due to security issues in the Casper protocol. Here is a twitter discussion:
Comparison No.3: Polkadot Grandpa
A new BFT finality gadget protocol inspired by the Casper FFG. A validator votes for a block on the tree, and a block gets some X number of votes when X sums up both this block’s and its descendent blocks’ votes.
Security issues in Grandpa: Malicious groups can launch an attack and lead the Polkadot’s blockchain system into a deadlock.
Assume that Er-1,0 = … = Er-1, n-1. During round r, the block production mechanisms produced a fork for Er-1,0. That is, two child blocks B and C of Er-1,0 are produced. At round r, t + 1 voters (including all malicious voters) prevote for B and the remaining honest 2t voters prevote for C. For each voter Pi, we have g(Vr;i) = Er-1,i. Thus each Pi precommits g(Vr;i) = Er-1,i. Now each voter Pi estimates Er;i = g(Vr;i) = Er-1,i. Since it is possible for Cr;i to have a supermajority for any child of Er-1,i, the round r is not completable. That is, the process stuck at round r forever.
Thank you for reading all the way to the bottom. Mark your calendar for our next virtual event in the last week of June to Ask Us Anything! Keep an eye out for the invitation across our social channels.
Meanwhile, Sperax South Korea is recruiting ambassadors! This is a great opportunity to get involved with Sperax team building, share the latest information and make friends. Please refer to the following rules:
Qualification: All the investors in the Sperax Korean community, early supporters, tech enthusiasts with dev background, community opinion leaders. We welcome anyone who trusts Sperax and wants to grow up with us!
Recruitment time: 2020.05.26–2020.06.26
Ambassadorial duties:
Sperax Korean community expansion
Promote Sperax on various online communities
Application channel: Add group manager @Jennifer1003 on Telegram.
Rewards: We offer USD 200- 500 worth of tokens per month, depending on your accomplishments.
About Sperax
Founded in 2018, Sperax is a blockchain company focused on building a trusted infrastructure to enable decentralized financial services and create a decentralized economy. We have global talents working in Silicon Valley, Beijing and Singapore. Our headquarters is located in Menlo Park, CA, the heart of Silicon Valley.
Contact:
Website: https://sperax.io/
Twitter: https://twitter.com/sperax_io
Twitter Korea: https://twitter.com/SperaxKorea
Marketing: https://twitter.com/FridaCai96
Telegram: https://t.me/sperax
Telegram Korea: https://t.me/speraxkorean
Wechat微信公众号: Sperax
Blogs: https://medium.com/sperax
LinkedIn: https://www.linkedin.com/company/sperax/
Facebook: https://www.facebook.com/sperax/
Email: contact@sperax.io