From Visa to the Raiden and Lightning Networks.

Ethereum is genius, but slower than Visa...

From the very beginning, the main purpose of blockchains is secure, online peer-to-peer transactions from one part to another, without the need of an intermediary like a financial institution. But in order to become the transactional infrastructure of the future economy, blockchain-based transactional systems also need to be fast and cheap.

What makes it slow

In all blockchain protocols each node stores all states (account balances, contract code and storage, etc.) and processes all transactions. This provides a large amount of security, but greatly limits scalability. A blockchain cannot process more transactions than a single node can. Because of this, Bitcoin is limited to ~3–7 transactions per second, Ethereum to 7–15, etc. To put it into perspective, credit card companies process a growing number of transactions, currently more than 10,000 per second.

Are there ways to create a new architecture so we can achieve higher speed without compromising security and cost?

Scaling Ethereum to Billions of Users

There are a few Ethereum scaling projects in the works, each of which tackles a different scalability problem: Sharding, proof of stake, Plasma, and theRaiden Network (lightning) among others.

Today, I will focus on the Raiden Network.

The Raiden Network

One of the most promising approaches to make blockchains scalable is the off-chain state channel.

Imagine Persons A and B, who want to make a transaction on the Ethereum blockchain.

Instead of doing the whole transaction on the Ethereum blockchain and relying on the speed and scalability of it, what they could do is create a peer to peer channel on-chain but do the transactions of the channel off-chain, apart from the last one (the settlement).

A and B have to pre-define an amount of Ethereum tokens they will put on this so-called channel. After they define the deposit, they can send multiple signed transactions between them in both directions off-chain, while a ledger between them is updated with the latest state.

When the two parties decide to close the channel, only the final state of the transactions between the two is broadcasted to the blockchain. In other words, the blockchain is only used to open and close the channels.

Both have to sign (multisig) when to get the funds out, close the channel, and get the last state out on the blockchain (broadcast).

Also, the Raiden Network (same for lightning) supports routing of transactions over multiple hops. This means that user A can send a payment via a node C to B, if payment channels are set up between A and B, and between B and C.

To achieve this, A sends a transfer to B through C, and locks it with a secret, generating a hashlocked transfer. He also puts an expiry on the lock (time lock) in case C goes offline or proves to be malicious in other ways so that in the end A has the possibility to know that after a particular block his transfer won’t be valid.

The RDN Token

In contrast to the Lightning network, Raiden has a token.

The token will be used only for payment for peripheral services around the Raiden network. Such as third party channel monitoring. It won’t be required for the core part of the network functionality.

There is not a lot of information about the design and usage of the token. I am in contact with the team so they can help with more info regarding the token design.

Benefits

Faster transactions Users can make multiple transactions, and the confirmation takes a fraction of a second.

Network effect On the Raiden network every participant (node) only has to open a few channels, but will still able to make connections and transfer to any other participant (node). This means that the bigger the Raiden Network becomes, the more participants (nodes), and the more connections between the nodes, making it easier and faster to find a channel and make a transaction.

Improving privacy Ethereum transactions are public, whereas Raiden transfers will be private between the payer, the payee, and the nodes forwarding the transfer. When channels are settled, only the sum of transactions will become visible to the entire world.

Lower fees Raiden Network transfer fees will be small and off-chain, but lower than on-chain transaction fees. Instead of paying for global consensus, you only pay for forwarding peer-to-peer consensus.

Challenges

Lock up tokens All current payment channel protocol proposals require an initial deposit by both parties setting up the channel. This technical requirement results in the capital stake at the opening of a payment channel staying the same for the duration of the lifetime of the channel.

Low value transfers only The Raiden Network is not intended to support large value transfers. This is because nodes need to be equipped with enough deposit (see above) at setup time to route a transaction of a certain size through a channel. For example, if two parties want to transfer 1000 eth on the Raiden network, each node on the route should have at least 1000 eth of deposit at the setup time. The probability for this happening is very low, making it impossible to use Raiden for large-value transfers.

Leads to a centralized network It might very well be possible that wealthy institutions will create larger nodes through high deposits to make a profit on transfer fees, isolating the smaller nodes from the network.

Potential

Raiden is a great system for users who want to make frequent low-value transactions. Some use cases of the Raiden Network could be micropayments for low-priced digital goods, content, transfer tokens or credits between games, or even decentralized exchanges of digital assets.

The necessity of Raiden Network to the ecosystem is equal to projects like Polkadot and Plasma, and it’s planned to be deployed on the Ethereum main net.