QUARKCHAIN – A HIGH-CAPACITY PEER-TO-PEER TRANSACTIONAL SYSTEM

Hello! My name is Rose and I am new to Steemit! I am passionate about technology and I’ve been in the cryptocurrency world for about a year now. On this blog, I am going to start doing reviews about cryptocurrency projects that catch my interest, as a way to learn, and also to help others! I hope people enjoy it!

So, as many of you may have noticed, we recently have had many new crypto projects that claim to be the next big thing when it comes to blockchain generation 3.0. In this post I am going to explain the Quarkchain’s vision and how they are going to achieve their goals. But first let's take a look at what are some approaches to scalabiliy that some competitors are offering.

So, what are the main approaches to the scalability issue today?

One of them is the Lightning Network:

The basic idea of the Lightning Network is to defer frequent transactions among a fixed group of parties until all parties are finalized with the transactions. Then, one of the parties would just post the final result without incurring multiple historical transactions on chain. A lightning network generally requires two transactions to create/destroy a payment channel, which accepts off-chain transactions. The number of off-chain TPS could be infinite in theory. However, the Lightning Network is only suitable for frequent transactions among a fixed group of parties, while it is inefficient if a user's transaction target is random and happens sporadically. Transparency is another concern because transactions are tracked through lightning channels rather than the main blockchain. Some off-chain solutions rely on trusted third parties, such as Paypal or Alipay with blockchain features. This prompts us to ask if it is necessary to build another centralized payment method because there are already many out there.

Another one is Sharding:

Sharding refers to horizontal partition of data to break a database into smaller parts. It is one of the most common ways in centralized systems to address the scalability problem. The main challenges for sharding include cross-shard transactions, security issues like single shard take-over, and further scalability issues. In this article, you are going to understand why Quarkchain’s implementation of sharding is superior in comparison with the competition.

Now, let’s analyze Quarkchain’s solution:

As stated in Quarkchain’s whitepaper, the three main challenges of a blockchain are: security, decentralization, and scaling-out. And although security, decentralization, and scalability are all important for blockchain, there are some tradeoffs among them. If one wants to increase the security/privacy, a larger amount of data is needed for each transaction. This means lower transaction speed and larger storage. Since in the last few years the demands in the blockchain space have increased tremendously, the ultimate goal of Quarkchain is to extend the scalability as high as possible while keeping security and decentralization in an appropriate level.

So, how does Quarkchain improves scalability?

Unlike many existing approaches that attempt to address the scalability problem by enhancing existing systems, QuarkChain is designed for scalability from the beginning.
Quarkchain has a different architecture than the other blockchains that are trying to increase the blockchain’s throughput. The idea is going to have two different layers, one being responsible for the scalability, while the other is going to guarantee the security.
One layer contains a list of minor blockchains (shards). Each shard processes a sub-set of all transactions independently, therefore, as the number of shards increases, shards can process more transactions concurrently. As a result, the system capacity increases as the number of shards increases.

The other layer, which is the rootchain, will confirms all blocks from sharded blockchains. The root blockchain does not process any transactions (since it is not economically efficient), but its block has sufficiently strong difficulty so that reverting any transaction, i.e., the transactions in root blockchain, is not economically efficient.

QuarkChain network is also designed to support additional shards in an active network. Adding more shards is easy and fast, while users barely sense this (the users may feel faster processing of transactions if the network is congested before adding shards).

How does Quarkchain solves the security problem?

QuarkChain will have around 50% (this number is reconfigurable) hash power allocation on the root chain, which is the layer responsible for confirming all the blocks from sharded blockchains (responsible for security). So, all transactions in the QuarkChain network are going to be protected by 50% of the overall hash power of the network, which means that a double-spending attack requires at least 25% hash power (50% * 50% = 25%). Even though this is smaller than single-blockchain's 50%, QuarkChain is more decentralized than their competitors are, thus a miner will be much harder to collect so much hash power in this network than on a single-blockchain.

Finally, how does Quarkchain solves the decentralization problem?

Traditional blockchain technologies require every node to fully validate all blocks and reject any block that is invalid. Similarly, the node in QuarkChain that validates all minor blocks and root chain blocks is called super-full node. If every node in QuarkChain runs as super-full node, QuarkChain could have the same safety level as traditional blockchains.
However, running a super-full node could be very expensive in a high-throughput blockchain system. For example, 1M TPS with each transaction being 250 bytes would require 2 GBps network bandwidth, which becomes a huge barrier to many users.

QuarkChain addresses this problem by allowing multiple honest nodes in a cluster to run as a super-full node. Each node in the cluster only validates a sub-set of chains. As long as the union of their sub-sets cover root blockchain and minor blockchains, they are able to fully validate the whole blockchains without acquiring an expensive machine. In addition, if one of the nodes crashes in the cluster, the rest nodes are still able to fully validate any blocks since any two of them form another cluster, enabling high availability of such clusters.

Furthermore, to encourage forming such clusters in the network, QuarkChain will have incentives for miners to answer a puzzle about the information of random blocks. The puzzle will perform over a large amount of blocks and it is memory or storage intensive, and thus downloading the random blocks on-demand from the network will be inefficient.

That’s all for this post!

Quarkchain is a very complex project and if you got interested, I would definitely recommend that you read the entire whitepaper for a more profound understanding! In addition, there are many other interesting aspects that I choose not to include in this article. But maybe in another one!

Thanks for reading!