Led by Professor Emin Gün Sirer of Cornell University, one of the cryptosphere’s most esteemed critical thinkers, and the co-founders of AVA Labs, Kevin Sekniqi, CS Ph.D. at Cornell and Ted Yin, CS Ph.D. at Cornell, the “AVA Labs” team recently revealed the concept behind “Alpine Snowstorm”.
This took the form of a private testnet for a next-generation blockchain platform based on a novel metastable consensus protocol family called Snow-Avalanche.
However, does this represent the next chapter in the story of blockchain technology?
Very likely, but let’s try to understand why.
Despite several projects having emerged over the last two to three years, there has not yet been any major disruptor in the blockchain space. Although we constantly see projects adding features and building castles in mud fields, this progress is hamstrung by serious tech bottlenecks that are rarely addressed successfully.
As well as these, there are many other well-known projects that have directly inherited some of the major problems inherent to blockchain 1.0. To combat these, a wide variety of tools have been introduced in Blockchain 2.0, which improve scalability, performance, and functionality. These incorporate additional layers such as the Lightning Network and State Channels both of which are side layers rather than layer 1 fundamental solutions. These often involve trade-offs where, in most cases, decentralization falls victim.
What is proposed here is a metastable consensus protocol inspired by epidemic protocols and gossip networks which comes with lessons learned from Classical Consensus and Nakamoto Consensus. This serves to combine the best of both to fundamentally improve the known issues in layer 1.
The Whitepaper released by Team Rocket on IPFS in May 2018
This paper introduces a family of leaderless Byzantine fault tolerance protocols, built around a metastable mechanism via network subsampling. These protocols provide a strong probabilistic safety guarantee in the presence of Byzantine adversaries while their concurrent and leaderless nature enables them to achieve high throughput and scalability
Rather than create additional layers of protocols or sacrifice X for Y*
Avalanche introduces scalability, quick finality, extreme performance, and decentralization on layer 1. Moreover, the team will take the opportunity to apply lessons learned from blockchain 1.0 and 2.0 in order to solve major problems that are barricading the road to wider adoption.
*Decentralization for Performance; Scalability for Security; Decentralization for Security; Scalability for Performance;
Besides the fact that AVA benchmarked nearly 4 times higher the TPS of VISA, there is not much more to say.
Bitcoin — 7 transactions per second
Ethereum — 15 transactions per second
Ripple — 1500 transactions per second
VISA — 1700 transactions per second
PayPal — 193 transactions per second
AVA — 6500 transactions per second
What do we know about DApps today besides their lack of widespread adoption?
Although they certainly have various advantages over Web2.0 Apps, they are expensive to adopt. In a freemium digital world where users are accustomed to benefiting from technology and only paying for premium features, setting the bar as high as US$3 for breeding a virtual kitty or sending an encrypted message is excessive.
Such high adoption costs at the initial stages of innovation are nothing new. History is littered with examples of innovative products, including mobile phones, computer and mobile operating systems, and the internet, that initially were not accessible to the majority of users due to the complexity and costs of using them. However, as the technology improved and costs dropped, adoption grew.
In his 1991 book Crossing the Chasm, Geoffrey Moore argued that the key to achieving breakthrough adoption for high-tech innovations was overcoming the “chasm” in order to reach the “early majority” (the pragmatists). He describes this chasm as the massive gap that lies between the early adopters (tech enthusiasts and visionaries) and the early majority, which exists when a new product has the potential to be highly disruptive and thus require behavioral changes — James Kilroe and Seamus Hennessy
Source: (Moore, 1991; Nesmith, 2018)
Besides the cost of adoption, DApp performance is very limited owing to the long period of time required for a transaction to enter a block.
Example: I have an encrypted chat DApp. It makes a lot of sense to utilize blockchain tech because it is decentralized, immutable and has no single point of failure.
At today’s parameters, each message will cost users US$ 0.125 and the message will be received after 4–29 sec has elapsed (13 sec on average) If the network is under a very high load which is not hard to occur the users might have to pay up to US$3–4 per message and wait for it even longer as blocks would be full. Although I appreciate the benefits of blockchain tech for my DApp, it’s slow and expensive.
The encrypted messages will require between 1–2s to reach its recipient.
The TX cost will be low for two reasons: (i) flexibility to control their cost value; and (ii) the network cannot be easily overloaded. Now my DApp starts to make a lot more sense.
In essence, Avalanche is a crowd oracle, where economical parameters can be voted on and adjusted dynamically without the need of a hard fork.
That means that at any given moment the network participants can vote on the platform’s best interests and change the costs for certain transaction types.
AVA is the platform of platforms. It allows all kinds of digital assets to be issued by anyone. What makes it different and special is that AVA will allow multiple scripting languages and multiple virtual machines. This means support for a variety of nodes with different capabilities, opening up a new spectrum of digital assets, functionalities, and capabilities in the process.
It is a construction kit, or LEGO if you like, for digital assets! — Prof. Emin Gun Sirer
We like to think we are experiencing decentralization and the chains are completely immutable when we are using top cryptocurrencies based on Nakamoto Consensus at the expense of massive amounts of energy and expensive hardware, but what is the reality?
Bitcoin is held by many people, and several others support the network by operating a full node, yet it is a sad truth that bitcoin mining and therefore the bitcoin ledger is in the hands of 13 mining pools:
Only 4 of those are enough to reach consensus.
Ethereum Mining Pool’s Hashrate Distribution
The truth is no different. 2–3 mining pools are able to reach a consensus and supposedly alter the chain or basically endanger its immutability.
Avalanche introduces extreme decentralization which enables very high immutability in the ecosystem by allowing thousands to millions of network participants by being lightweight, easy to implement/understand and has no specialized hardware requirement.
AVA introduces economic governance through voting, a feature that rests on the very core of the protocol itself. The consensus is reached through subsampled voting, and the same voting mechanism can be used for completely decentralized governance. This allows changes to be made on the protocol level, including critical parameter adjustments.
By essentially being a crowd oracle AVA proves to be extremely flexible without risking the ledger’s immutability.
In other words, no more hard forks.
As expected at this stage, full details regarding UX are lacking. Nevertheless, the AVA Labs Team has acknowledged the problem with user experience and the obstacles which it poses for further adoption.
Advancing UI/UX was also on Prof. Sirer’s agenda at the NY TokenSummit 2019.
Revolutionary is the one word I have for the work of Team Rocket and AVA Labs. The combination of protocol capabilities and the knowledge acquired through the past years of experience and research gives the right tools in the hands of the team to craft the ultimate blockchain platform of platforms.
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —
1/ Snowflake to Avalanche: A Novel Metastable Consensus Protocol — Team Rocket 2018
2/ Scalable and Probabilistic Leaderless BFT Consensus through Metastability — Team Rocket 2019
3/ Emin Gün Sirer - MONEYCONF 2018
4/ Emin Gün Sirer - TOKENSUMMIT 2019
5/ Kevin Sekniqi - Neutrino 2019
6/ James Kilroe and Seamus Hennessy — Bitcoin Magazine 2018
7/ Mohamed ElSeidy - Demystifying “Snowflake to Avalanche”