TRON 4.0 Online Conference: Panel Discussion on The Rise of BTFS Distributed Storage

In addition to Dlive, another star project in the TRON family is BTFS. BTFS is both a protocol and network application that provides a content adjustable, peer, to peer mechanism for sharing content in a decentralized system. This is known as DFs. BTFS ecosystem consists of five components, namely BTFS network, TRON network, BitTorrent network, TRON, and BTE.

As a crucial part of the BitTorrent ecosystem, they are engaged in sound interaction with the BitTorrent speed. We like other products to empower users. Now, let us join Charles, CEO of Dlive, Justin, head of BitTorrent head of product engineering and Eric,s technical lead on BTFS on the rise of distributed I have storage.

So let’s go to the next question. The distribution network, one of the core Blockchain technologies is to have the key growing field for the future. The birth of the distributed storages as — into the distributed network applications. Although lots of companies have delved into the distributed storage since the rise of the internet, a few of them have delivered actual breakthroughs.

Many of us are familiar with the history of distributed storage. So would you please share with us your insights on this subject? So Charles, please go first. Thank you!

>> Charles: Okay. I will do it first. So if you’re asking this question a couple of years ago, I would say it’s more like a concept, right? At least for video platforms. Two years ago in the beta version of Dlive, when we were launched on the steam blockchain, Dlive had a building, a video playing support for uploading videos in the past broadcast.

We try to implement IPFS to host the videos but it didn’t work out well. Most of our user experience lagging or sometimes the video isn’t even playable. Now, with the solution of the BTFS provided, we are able to see a better result in terms of user experience. That means everything, you know?

Especially to app developers like Dlive, we can tell that the distributed storage services have been developed really fast in the past two years.

>> Moderator: Yes, so Justin, can you go next?

>> Justin: Sure! Yeah, so as Charles said, I think if you asked a few years ago, I would have said that distributed storage is an idea ahead of its time. The project in this space had been around for many years but usually with pretty limited success. The reason is that making it work is pretty hard when you consider challenges like network return where hosts are leaving and entering the network very frequently.

Bandwidth asymmetry where you upload might be much smaller than your download rate and the ability of people to free ride on the system or like to see you and say, I have stored these things and in fact, they had not.

The solutions that had been proposed have been to solve network turn by adding redundancy or adding correction codes, like, BTFS does, you can break the file in small pieces, distribute them broadly, and have codes to reconstruct the file if something is lost.

That’s a pretty good algorithm trick to deal with network return. It can also help with bandwidth asymmetry because you can be sure with some probability that some of the hosts that have pieces will have good up links or you can use the network upstream connectivity as a signal in terms of how to distribute those chunks.

And then the last problem is around free riding or deception. That I think is the piece that fell into place at the latest. So I think the solution there with blockchain and cryptocurrencies is you can create incentive systems where you don’t have to rely on people to store things. You can make it so there’s an economic incentive for them to do that.

In terms of storage deception, there’s also been a lot of progress coming up with proofs like proof of data position, proof of retrievability, proof of replication. All of these things make it much more difficult for the bad actor to poison the well by claiming to store something, earning tokens or coins, and not living up to the promise.

>> Moderator: Great point. Eric, it’s your turn.

>> Eric: So this is a great question. I believe many projects in the past have attempted to solve the fundamentals of distributed networking and even build for functional distributed storage systems.

We have gone through the initial breakthroughs of the fundamental principles and now are evolving to the area of decentralized and incentivize distributed storage space. The so similarly, Bitcoin wasn’t invented from day one. It took many years of research and effort into the fundamentals of E-money to materialized before brought in technologies were born and being utilized in everyday life.

So what I believe in the next steps of the centralized storage is actually fair and incentivize the network in a space so you can push it to the user adoption space and market.

So BTFS is such a solution to provide more flexible, cheaper, safer, and provides more privacy than traditional and even competitor products on the market. So I believe that BTFS has combined the information, distributed fundamentals, and Blockchain technologies to achieve these goals and more exciting features and breakthroughs to be built on BTFS in my opinion.

>> Yes, I believe so. And looking forward to that as well. So from the perspective of operations, product and media respectively, I was wondering what’s the rule of distributed storage space in the Blockchain industry and what is our internal and external reasons for the birth of distributed storage? So we’ll still go first with Charles, please.

>> Charles: Okay. No problem, so I think with Blockchain technology, distributed storage becomes a solution to many app developers. Decentralization is the most important mission that app developers with Blockchain technology can achieve.

When it applies to storage solutions, no centralized party shall be able to control the ownership of it, except for the owner of the content itself.

Before the distributed storage solution was invented to some users, having full ownership of the content is only possible by storing the content in the hard drive that they own.

So you can see there is a need in the market for distributed storage which is, I think the reason for the birth of the distributed storage solution.

>> Moderator: Yes, thanks for the answer. So Justin, would you like to share your idea?

>> Justin: Yeah, sure! I think that Blockchains are very great tools for distributed consensus over a small data set or medium-sized data sets like ledgers and things like that. They don’t scale well in large volumes of data like high-resolution video and things like this.

So if you don’t have distributed storage, then as a DApp developer, you are left with this dilemma. Then I can either have a pure decentralized solution but where I can store very limited amounts of data which basically rules out the entire categories of application. You can’t have a purely decentralized, you know, video app if you have to store it all on-chain.

Or you have to compromise and say, this part is pure and decentralized and this other part of it is going to compromise it and centralize. That compromise means that you’re vulnerable to attack. A single point of failure, client-server sites can be blocked so you end up with something that is like, the original intent of Blockchains.

So I think the decentralized solutions like redistributive storage allows you to build all of the classes of apps that you may want to, without making that compromise.

>> Moderator: Yeah, so, Eric, what is your thought?

Eric: I think this is an important product for Blockchain products. Echoing what Justin said, I believe that Blockchain storage is expensive does it must be stored on decentralized networks. Inoperability between Blockchain data and distributive data is also very crucial in building real-world products.

So any centralized components must be carefully avoided to keep Blockchain technology decentralized. Some I believe as the data sets grow and become more manageable, and more unmanageable, which is constantly discussed topic, the reasons for the distributed storage is similar internally and externally to an extent which is to solve the problem of expensive data and constantly growing data sets.

Also, I believe in additional, internally it must adapt components and externally, it helps real consumers save when they approach the Blockchain products.

>> Moderator: Yes, great point. So more and more companies are pulling efforts into the field of distributed storage. It is showing great potential and many have high hopes for it. So my third question is can you share with us the features and the advantage of distributed storage?

What are the core competencies that you think distributed storage should have? We’ll go first with Charles, thanks.

>> Charles: In my opinion, I think the most important advantage of distributed storage is prolonging the lifetime of the content at a much lower cost. For the video app developers, we know how costly it can be, an infrastructure where you have thousands of new content pieces being created every day for the platform.

>> Moderator: Thank you, Charles. Justin, would you like to share your opinion as well?

>> Justin: Yes, I think that distributed storage like Charles said, enables use cases where you can store large assets that are not necessarily going to be expensive to store and they are not going to be as vulnerable meaning they won’t degrade over time, information is lost from the internet, you don’t have version control and things like that.

So I think that key competencies include, persistence, versioning is important. The ability to understand what something was yesterday and what it is today. And both of those are important in terms of resistance to like adversaries like anticensorship as well.

I guess I would also compare it to HTTP. So if you look at the internet today, there are many protocols that you can use for storing and retrieving data. If you look at it from the proportion of traffic, the winner is HTTP by a landslide.

The web is the killer app of the internet but HTTP has problems including the fact, it’s centralized, it’s not content addressed. Anyone can DDOS. People can censer your sites so you have an accidental loss of data. So that is the gap that I think distributed storage needs to step in to fill.

>> Moderator: Yes, I agree. Eric, would you like to answer the third question?

>> Eric: Sure! So I think there are many parts. For example, security, privacy, infinite scalability, community governance just to name a few important features BTFS is looking to support and has been installing in certain categories.

One of the biggest advantages in my mind is still the ownership of data. So most traditional solutions do not provide enough privacy for your data and it could be sold to any party without you knowing it, although there are laws governing it, you don’t know what is happening behind the scenes.

So distributed storage is not only fair but gives the users the power to control, how, when, and how much to store on the network, anonymously and safely.

So second to that, the second advantage is availability. So as Charles mentioned, traditional technology is usually relying on central clusters or data centers that are prone to downtime and failure if a provider suffers an outage or human faults from the company or the team running the operations.

So high availability and mass backups are crucial competencies in the distributed storage space which I believe BTFS will be the leader in the space.

>> Moderator: Very thought-provoking. Thank you. This is a distributed storage that serves as a very foundation of how an entire Blockchain world is built and is bound to face higher market demands. We expect a stronger plan for BTFS to be applied in many business app scenarios moving forward. Thanks, guys!

The official Medium of Tron Foundation. Learn more: https://tron.network