Birth of the Avalanche Protocol : Interview with Emin Gün Sirer Co-founder and CEO at Ava Labs

Emin Gün Sirer (Co-founder and CEO at Ava Labs)

Interview Date : 28th July 2020

Interview with Emin Gün Sirer

Cornell University Professor Taking on P2P and Consensus in 2003

I am a professor at Cornell University. I have been a professor for 19 years now, and I started looking into cryptocurrencies and peer-to-peer systems, especially for file-sharing, in 2001.

In file-sharing, there was a problem where people who downloaded files would not upload – people who were called “leeches” as they would take but wouldn’t give. This happens to be a common problem for most P2P systems.

My idea back then was to have a decentralized currency where you have to pay for downloading, and are rewarded for uploading things with a digital currency. In that way, people would ultimately run out of coins and have to contribute by uploading new material. This idea gave birth to a system called Karma that used Dr. Cynthia Dwork’s proof-of-work approach.

I published Karma in 2003, about 6 years before Satoshi and Bitcoin. Unfortunately, I was in the US and Karma was in its earliest stages just after 9/11. I was a young professor, and my colleagues told me that what I was creating was really interesting, but that I would never get funding for this kind of work as there’s so much concern about terrorist financing.

They were right, and I don’t think I could have built a career on decentralized currencies back then, so I stopped working on Karma and moved on to new areas of study.

Finding “Selfish Mining” in Bitcoin

I came back to cryptocurrencies when Bitcoin came along. While I recognized the breakthrough Satoshi had created, I discovered the biggest flaw in the Bitcoin protocol, a problem called “selfish mining” where miners could collude to increase their revenue and centralize operations.

People initially denied its existence, and crowdfunded a whole bunch of studies to try and prove us wrong, but eventually concluded our conclusions were exactly right and became our best partners.

I started getting invited to all kinds of conferences and got a reputation as the person who found the biggest known flaw in Satoshi’s work. I began working on coin security and making the coins more scalable by working on a protocol called Bitcoin NG ( Bitcoin Next Generation) as well as  Bitcoin covenants protocols that were very geeky, very detailed improvements to various protocols.

Different Consensus Approaches are Needed

Consensus protocols are the foundation and most important characteristic of these systems, but there hadn’t been a breakthrough since Satoshi’s consensus protocol.

When I look around, I see at least thousands of different independent projects listed on CoinMarketCap, and yet the vast majority of those projects don’t bring any new innovation to the table. They slightly modify existing, well-researched classical protocols built on the amazing work of American computer scientists Leslee Lamport and Professor Barbara Liskov and present their project as a breakthrough.

Nakamoto consensus represents the top of what we know today, and in those 10 years, everything else that came afterward, has either just been a re-creation of Satoshi’s great ideas, or going back into history, they are trying to resurrect all the protocols that we all knew about, and that Satoshi knew about and didn’t believe were suitable for an open internet–and he is right.

To the public, it looks like a unique project, because it looks a little different and has a different management team behind it. Perhaps, it’s got a different focus as well. But most of these proof-of-stake systems are essentially recycling all the academic results from the 1980s~1990s.

About 2 years ago, however, we had our first major breakthrough since Satoshi with the Avalanche protocol. It represents a genuinely new approach to consensus protocols, and is only the third new approach in 45 years of research in distributed systems, after Classical consensus and Nakamoto consensus.

Ava Labs – New Innovations in Consensus

I am heading Ava Labs, the company behind the Avalanche platform.

We are a company of 40 to 50 people at this point making it simple to launch finance applications using blockchain technology–with highly scalable and efficient networks, customizable public and private blockchains, the capability to create any digital asset, and more.

Our goal is to innovate at every layer of blockchain networks. Beginning with a breakthrough in consensus protocols, and continuing to areas that no one had been looking at like the network and virtual machine models evolutions that were obviously the next step for crypto, and yet, for some reason no project decided to take it on.

Avalanche achieves sub-second transaction finality, supports Visa-level transaction throughput, is resilient to 51% attacks, and is the only coin that supports multiple virtual machines.

We are going to show the world what exactly blockchains are capable of.

Birth of the Avalanche Protocol

It all started from this paper that we received from a team known as Team Rocket, who had been working on novel consensus protocol.

My students and I were already exploring lighter weight protocols than Nakamoto and Classical consensus to find unique approaches nobody had explored before, but that could still provide the same guarantees as Bitcoin. We reviewed the paper draft from Team Rocket, and quickly realized that we were on to a big breakthrough.

Avalanche didn’t rely on proof-of-work or mining, but used probabilistic consensus based on sub-sampling. It could be secure and predictable like Bitcoin, but didn’t suffer the centralization or performance trade-off inherent in Classical consensus approaches.

When we had these results in our hands, we knew we had to see it through completion and build a full system to showcase Avalanche to the world.

Avalanche: Sub-second Finality, VISA-Level Throughput, Resilient to 51% Attacks, and Scalable to Millions of Validator Nodes.

One big difference between Avalanche and everything else is the consensus protocol. There is this conception that people have acquired over time, that blockchains are slow and not scalable – and that has proven true with existing blockchains that are very slow and do not scale.

With the Avalanche protocol, we use a completely novel approach to consensus that can achieve the same guarantees of safety, while achieving very quick finality, high-throughput, and meaningful decentralization.

For example:

• Bitcoin achieves transaction finality in 1 hour. Avalanche transactions achieve finality in about a second.
• Bitcoin can currently handle 2 dozen miners. In Avalanche there could be 10 thousand to many millions of miners or miner pools without leaking value out of the system.
• Bitcoin can facilitate ~4.6 transactions per second. Avalanche has achieved 6500 transactions per second, 3 times more than VISA.

The best part is that Avalanche achieves these performance numbers without compromising decentralization – a common approach by projects looking to inflate their numbers. Avalanche is more decentralized than existing blockchains, reaching over 1,000 full, block-producing validators in the system, and yet we are still producing speeds in excess of VISA.

Honestly, in all of my time researching these systems, I never thought a decentralized system could achieve those performance figures. It took the breakthrough of Avalanche consensus to make the impossible possible.

Avalanche: Multiple Virtual Machines, Customizable by Subnet

Unlike other networks that force terms and conditions of network participation uniformly across the system, Avalanche empowers individuals and enterprises to determine what rulesets and economic incentives govern their use cases.

Take, for example a regulated financial institution wanting to issue digital assets. It’s simply not possible for them to do so compliantly in a system like Ethereum where they cannot control which nodes validate their network activity. They have one choice of virtual machine, and no choice over virtual machine or network parameters.

In Avalanche, the same institution has complete control over the virtual machine running their use case (Ethereum Virtual Machine is completely compatible with Avalanche) and the conditions of running a node on their custom sub-network (e.g. regulated by same authority, in the same jurisdiction, operating requirements).

Institutions have invested billions of dollars in financial technologies looking for every edge in performance to outpace their competitors, but have been told by blockchain projects that they now have to settle for subpar performance and relinquish control to achieve these magic efficiencies.

That argument doesn’t hold up under scrutiny, and I suspect that’s the source of financial institutions becoming disillusioned with blockchain technology. They simply didn’t have compelling business reasons to offset the risks of migrating to emergent systems.

We hope they’ll see the difference Avalanche delivers in performance, security, control, and flexibility to make their systems compliant by design and efficiently up to hundreds of thousand of participants. The end result is a market structure defined by velocity, efficiency, and innovation in new products and services. 

Outlook for Cryptocurrencies and Blockchains

I am very bullish on crypto.

The World’s economies have lost their footing due to the COVID-19 pandemic and the subsequent injection of easy money into the system. Almost all assets are being artificially inflated to guard against the adverse effects of a recession.

Today, the U.S. stock market and real estate sector are completely bloated beyond belief. Commercial real estate is down, but wherever you look you see assets that are trending toward worthless or on life-support from monetary authorities.

Crypto has largely been an exception to the bubble and more of a passenger being dragged along for the ride. I expect this to continue for some time because there are so many people looking for a safe harbor.

Stocks are sound investments because by owning stock certificates, it can give you returns on future incomes. However, we are looking at the price to earnings ratios that are stratospheric. Simply put, people can only put so much money into the stock market before it becomes ludicrous and tips over on systemic risk.

We’re at the point where investors are asking themselves, do I really want to put yet another dollar into this insanely inflated market? I think the vast majority are realizing these marginal dollars are not going to yield the returns that they would like, and it’s just going to disappear into the hands of a team who doesn’t know what it’s doing.

Nevertheless, there also are some very worthy projects in the crypto and DeFi space that can provide great value. In particular, there are groups of people who have been using crypto as a way to jump-start new economies, projects, and innovation in the services available to the masses.

That’s a much more attractive – and reasonable – destination for people to put their money.

Interviewer , Editor : Lina Kamada

【Disclaimer】

The Article published on this our Homepage are only for the purpose of providing information. This is not intended as a solicitation for cryptocurrency trading. Also, this article is the author’s personal opinions, and this does not represent opinion for the Company BTCBOX co.,Ltd.