An overview of the QuarkChain Project

A successful trustless blockchain should have the following three features: it has to be secure, decentralized, and scalable.

It has to be noted that security and decentralization has be achieved up to a satisfactory level in the current blockchains. For example, Bitcoin blockchain is rather decentralized regarding its network (but it is rather centralized regarding mining).

Achieving two out of the three in a satisfactory level is rather easy, but achieving all three together is challenging, and it is the holy grail of the current blockchain technology. Actually, experience has shown that there are some trades off between them. If one wants to increase one of them, he might have to compromise regarding the other two. Therefore the question is to find the right balance between security, decentralisation and scalability. In other words we want to create a blockchain which is reasonably secure, reasonably decentralized, and scalable.

The most successful blockchains at the moment, like Bitcoin and Ethereum are rather secure, and rather decentralized, but they are not scalable. Bitcoin can achieve 6 TPS, Ethereum can achieve 15 TPS. But these speeds are orders of magnitude less compared to current payment systems like VISA (56,000 TPS) or Alipay (more than 200,000 TPS). Therefore if we need to use blockchain technology in a global payment system, we should be able to create a blockchain which can naturally scale up so it can handle such kind of speeds.

A blockchain may scale up in several ways:

1. Off-chain scalability;
2. On-chain scalability, vertically;
3. On-chain scalability, horizontally.

Now we discuss what are the pros and cons of three above types of scalability.

Off-chain scalability is used by using off-chain channels to carry out a big part of our transactions. This is used in the Bitcoin Lightning network or the Raiden network in Ethereum. The idea is, instead of recording every transaction on-chain, we may use another (possibly a centralized network) to sum up some transactions, and then record the final result on-chain. But using centralized networks on the top of a decentralized blockchain defies the very idea of the blockchain technology, and poses serious security risks regarding the security of these centralized networks. On the other hand on-chain transactions are more flexible, immutable and therefore more secure. Therefore on-chain scalability is desirable.

Now we discuss the different types of on-chain scalability, namely vertical and horizontal scalability. The basic idea behind vertical scalability is to increase the performance of a single blockchain. This can be achieved, for example, by decreasing the block interval, or increasing the block size, namely the amount of data that we include in a specific block. But there are computational and bandwidth limitations regarding this approach, and therefore there are restrictions on how much we can scale up.

Now we explore the basic idea of horizontal scalability. The basic idea behind horizontal scalability is the “divide and conquer” paradigm. Namely, we increase the number of nodes of the network, and we break the blockchain into smaller pieces. By doing so, we can achieve the following:

1. Since the size of blockchains is smaller, the processing time for an individual transactions is reduced;

2. The smaller pieces of the blockchain can be updated simultaneously and therefore we can process several transactions at the same time.

Quarkchain scales using both vertical and horizontal on-chain scalability. In this document we will focus mostly on horizontal on-chain scalability.

Now we see how, in general, we can implement horizontal scalability using sharding. We first define what the term “sharding” means. Sharding was first invented for databases (centralized ledgers), well before the introduction of blockchain technology. Sharding refers to breaking up a large database in smaller parts, so that we can scale up and increase our performance. It is based on the “divide and conquer” paradigm and it has been used successfully in centralized systems (BigTable and Cassandra are based on this idea).

Now we discuss how sharding can be implemented in blockchain technology. The idea is very similar to the sharding for databases. We partition the blockchain ledger in smaller parts, and therefore every network node is processing a much smaller part of the total ledger of all the system. As a result the speed is increasing. A super node in the quarkchain blockchain is defined to be a cluster of nodes, which can validate the whole ledger of the blockchain. Building super nodes by clustering nodes together has the following advantages:

1. Nodes do not have to be very demanding regarding CPU power and memory. Therefore the upfront cost for a potential miner decreases since he does not have to buy an expensive machine to be part of the node network.

2. It increases the speed and it is easy to scale up, since we can just add more shards to make the system faster.

The idea of using sharding for horizontal scalability has been already adopted by Ethereum, but it might take 3–5 years for Ethereum to complete this task. The reason is, it is much harder for an already existing blockchain like Ethereum to implement sharding but much easier for quarkchain which is brand new and made from scratch and designed specifically for scaling up. Some of the main challenges of applying sharding in a blockchain involves cross-sharding transactions and security issues (like one malicious shard to take control of the blockchain)

Quarkchain came up with a new type of blockchain which adopts the best features of the current blockchains, but it greatly expands on them. The basic design principles of Quarkchain blockchain are the following:

1. Enhance scalability while retaining security and decentralisation

2. Cross-shard transaction is effortless; The user can send and receive funds without being aware of the sharding architecture of the system.

3. Easy account management: The user needs only one secret key to access the block chain. This is in contrast to most of current sharding implementations where the user needs to hold a different key for every shard, which is not user friendly at all.

4. It is based on an open standard, which is compatible with EVM, and therefore existing Dapps can run on the Quarkchain without much effort.

5. It established an incentive-driven ecosystem. More specifically, the architecture of the system is designed so the hash power is distributed evenly between the shards. As a result mining is split evenly too. As a result the whole system speed (TPS) can increase as the number of shards increases and it also prevents any shard to take over the whole system.

Now we see an overview of the Quarkchain architecture. Quarkchain blockchain is based on a novel two-layer blockchain. In the top layer, there is a root-chain, and in the bottom layer there many smaller blockchains, which we call them shards. Each shard processes a subset of all transactions independently from each other, and therefore many transaction can be processed concurrently. The root-chain confirms the blocks from the shards. Therefore the shards are focused on processing blocks, and the root-chain are focused on confirming the blocks. Around 50% of the hashing power is devoted to the root chain, but it can vary from 25% to 75% depending on the circumstances.

Quarkchain runs a novel consensus which is hybrid of POW consensus. The details are as follows; The root-chain is running the standard POW consensus which is used by several other cryptocurrencies, including Bitcoin and Ethereum. According to this consensus algorithm, in case of a fork, the longer verified block survives. The shards, though, use a different consensus algorithm, which is a variation of POW algorithm, and which is called “Root-chain-first POW” consensus. According to this consensus, if there is a fork in a shard, them the fork with has the longer verified root-chain will survive, independent of the length of the forks within the shard.

Within such a framework, in order for a malicious agent to carry a double spend attack, he has to control the root-chain and the shards, and therefore it will require more than 50%*50%=25%.

Now we see how Quarkchain compares to other possible competitors. As we mentioned above, Ethereum is trying to scale up using sharing. But it turns out that it is much harder to scale up an existing blockchain, than creating a new blockchain from scratch, and therefore it might take Ethereum around 3–5 years before it scales up fully. Another blockchain which has addressed the scalability issue is Zilliqa. Zilliqua’s architecture is also based on the “divide and conquer” paradigm which is implementing by a form of sharding. But the Quarkchain smart contract language is Turing complete and Zilliqa’s smart contract language is not Turing complete. A given programming language is Turing complete, if we can express in the given programming language any possible program. Roughly speaking, Zilliqa does not provide infinite loops, and therefore if a smart contract requires to use infinite loops, then it might not be possible to run in Zilliqa. This makes Quarkchain more generic and powerful, and therefore better suited for real life applications.

Quarkchain blockchain can have a lot of applications in real life problems. As the CEO of the project explained, they plan to use their blockchain in many different areas, including but not limited to, finance, social media, and business enterprise.

Quarkchain blockchain will use QKC as it currency. This is considered to be a utility token. It will be mostly used as a way for the consumers to pay to run dapps on Quarckchain. So basically with QKC you can buy computation, similarly to ETH.

We conclude that Quarkchain is a big step in the blockchain technology which may change the blockchain landscape. The new blockchain Quarkchain introduces will give to different types of business the opportunity to operate in a blockchain set up. Therefore we might see new products and new services that have not been available before.

References:

1. “Quarkchain Founder chats with BCB about a Super Scalable, Enterprise Grade blockchain”: https://www.youtube.com/watch?v=42JWY3UWiqM

2. QuarkChain white paper: https://www.quarkchain.io/QUARK%20CHAIN%20Public%20Version%200.3.4.pdf