Blockchain Scalability Problems and Solutions
Blockchain is constantly growing since newly completed blocks are added to the chain of previously-formed blocks. These blocks are added and “saved” in a linear, chronological order. Due to the fully functional (anonymous) network of the blockchain system, a variety of advantages of blockchain technology for business opens up. If you have your own blockchain, there is no need for an intermediary to check transactions; instead, a consensus mechanism is used to verify transactions. However, despite the numerous advantages of blockchain technology, it also has serious technical problems. One of them is scalability.
Ethereum Plasma Overview
The Ethereum project strives to create a platform through which calculations will be conducted outside the blockchain, and eventually ensured in a chain that scales to billions of calculations per second with the least possible number of updates on the chain. With the help of Plasma technology, nodes will not have to confirm all data during the conclusion of the next smart contract. More precisely, the operating system, in this case, avoids the mandatory stages in loading the full history and verifying all information. An updated transaction will not reload the history of the block. The new transaction will require only a two-way exchange of data on the transfer of funds from the wallet of one user to the wallet of another. As a result, users get a much smaller amount of data that requires processing. From a technological point of view, a specific user will only have to monitor the accuracy and availability of his or her own coins that are engaged in the transaction.
Developers of the new Plasma subnetwork report that the use of their technology, which allows the implementation of transactions between trusted nodes without having to refer to the main block, can be implemented within a variety of projects. Cryptocurrency exchange platforms, blockchain platforms, and decentralized social networks can all use Plasma technology to increase the parameters of speed and security.
Plasma Ethereum protocol will have flexible settings that allow it to be tailored to the unique needs of a particular product. This will improve certain high-necessity aspects of the operation of a specific site. At the same time, the main function of Plasma – the implementation of more advantageous scaling and faster indicators – will not suffer. Each Plasma release can be configured in accordance with its business logic, based upon smart contracts. These can surround the root blockchain of Ethereum, solving the tasks of security and scaling. In the course of blockchain operations, some calculations will take place in child blockchains, resorting to the root blockchain only occasionally. However, there is the possibility of confronting local attacks. Members of the local side chain will have the opportunity to leave it by going to the root blockchain without losing personal funds.
Due to the side-chain users, Ethereum will build a tree-like architecture with indicators of branching and efficiency. This will help increase the maximum threshold of data security and availability while maintaining a minimum commission. Every sidechain can create its own token, which will be used as a reward for miners and for other purposes. The mining itself will be conducted according to proof-of-fraud protocol.
To understand the role Plasma can play in the Ethereum blockchain, we must explore the architecture of the network. The Lightning Network technology (as in Bitcoin) was adapted for Ethereum in the form of the Raiden protocol. Raiden is a kind of extension of Ethereum. For successful operation of processed transactions, nodes are linked both with the nodes of the root blockchain as well as with each other. Utilization of off-chain technology will reduce order commission since data exchange with Ethereum is rare regardless of the number of operations performed.
The use of Raiden will increase network bandwidth, but this is not the end of its benefits; the solution allows the use of smart contracts in a more utilitarian way. Here, the Plasma release enters into the game: a high-stability framework that allows the translation of certain computations from the root blockchain into subordinate units.
At the structural level, Plasma is presented in the form of an add-on from the set of smart contracts that are performed on the main network. It allows the processing of a large number of operations with very low main blockchain involvement in the data exchange. Thus, Plasma is an analog of a network of district courts, and Ethereum will become an analog of a federal court.
Separation of duties allows transference to each network only the data necessary to solve a specific task. However, for convenience, one has to pay: the danger of attack increases and security decreases. To eliminate fraud, a mechanism for imposing penalties for attempts to conduct illegitimate transactions was developed, in addition to the possibility of a quick exit from the subsystem to the main network.
In the Plasma release, the team proposes the possibility of creating its own tokens for each of the add-on networks (similar to the coins received by miners as a reward). This will create an economic incentive for validators, motivating them to maintain the operability and security of networks in accordance with the rules of fraud proofs (if a fraudster is discovered, he immediately loses his deposit, and the user gets the opportunity to use another network-add-on to make a payment). These instructions are the basis of the logic of the smart contract the prevention of illegal transactions. The withdrawal of funds takes time, which is used to check and cancel the operation if necessary.
Even if you send millions of transactions to the network, the commission will remain low, because to confirm the calculations, you will only need a few entries in the main blockchain (for example, if you send transactions once a day).
The processing performance of operations will increase substantially, because many small transactions will now be carried out outside of the blockchain, merging into larger transactions that enter the blockchain account via Raiden. The Lightning campaign can also be used to scale smart contracts and change their states. This is the essence of Plasma.
Scalable Blockchain Computation
In the Plasma project’s white paper, there is a detailed explanation of this novel technology. Plasma consists of five key components:
- a motivational ground for persistent contract calculation, which will be economically beneficial
- an architecture for organizing child chains in a tree-like shape for maximizing effectiveness at a lower cost
- MapReduce architecture for building Proof-of-Fraud for state transitions in inserted chains, which must be consistent with the tree architecture redrawing state transitions to higher scalability
- a consensus algorithm that depends on the parent blockchain, which tries to reproduce the outcomes of Nakamoto consensus drivers
- Bitmap-UTXO architecture to ensure precision state relocation from the parent blockchain, thereby reducing exit costs. Permission for an exit due to data inaccessibility or other Byzantine demeanor is the main concept of Plasma release.
Thus, the architecture of Plasma may be portrayed as follows: framing all blockchain calculation into a group of MapReduce frameworks, plus an additional way to carry out Proof-of-Stake token linkage ahead of preexisting blockchains, with the conception that Nakamoto consensus drivers disincline block restraint. This type of architecture is ensured by smart-contract development on the root blockchain utilizing a Proof-of-Fraud mechanism.
MapReduce is a framework for calculating/computing distributed tasks utilizing a large number of computers (nodes). We will portray this function in code form to demonstrate how they work.
The Ethereum team composes blockchains in a tree-like layout and processes each as a separate side-blockchain with imposed blockchain history and MapReducible calculations placed into Merkle proofs. By reframing one’s block into a child blockchain reinforced by the parent chain, users can ensure a great scale with minimized trust (supposing root blockchain presence and accuracy).
In the blockchain, the decision to ensure correctness usually depends upon each participant checking the chain him/herself. To accept a new block, one needs to fully test the block to ensure correctness. Many efforts to scale transactional throughput of the blockchain demand the utilization of temporary obligations to create a reliable bond (approval) such that claimed data must be subject to a controversial period allowing participants to ensure compliance with the state. This assert/challenge architecture helps determine whether the particular state is correct. If the value is wrong, then there is a controversial period when another observer can provide evidence challenging this statement within a certain decided time. In the case of fraudulent or erroneous demeanor, blockchain can punish (amerce) the guilty. This creates a mechanism for encouraging participants to impose a consequence, but if and only if the incorrect state is claimed. With this assert/challenge/proof architecture, interested participants can affirm ground truths for non-interested participants on the main blockchain.
This architecture can be utilized not only for payments but for calculation, making blockchain the decision-maker for contracts. However, the presumption is that all parties are engaged in verifying the calculations. For instance, in the Lightning Network, the design does this so one can set obligations with regard to the status of the smart contract (for example, with a pre-assigned tree of multi-signature transactions of the conditional state).
Externalized Multiparty Channels
The Ethereum team has proposed a method in which multi-party channels outside the network can hold state on behalf of others. The team named this structure the Plasma blockchain. For funds held in the Plasma chain, it allows deposit and withdrawal of funds into the Plasma blockchain, while state transitions are carried out with the help of Proof-of Fraud. This makes it possible to ensure operability and interchangeability since it allows withdrawing funds. The accounting of the Plasma block corresponds to the funds stored in the main chain (Plasma is not intended for compatibility with fractional reserve banking designs.)
A large number of transactions can be assigned in this chain with minimal data reaching the parent blockchain. Any participant can transfer money to anyone, including transfers to participants not in the existing set of participants. These transfers can withdraw funds into the parent blockchain’s token.
Ethereum Plasma allows the user (or a network of participants within a proof-of-stake network) to operate on the blockchain without a full permanent record in the root blockchain and without trust in a third party or parties. In the worst case, the funds are blocked, and the cost of time is lost with mass outputs on the blockchain.
Plasma Proof-of-Stake Consensus Mechanism
Plasma Proof-of-Stake allows interested parties to publish on the root blockchain or the parent Plasma chain, which comprises the assigned hash of their new block. The verifiers simply construct blocks that they have completely verified. They have the chance to create parallel blocks (to stimulate data exchange as much as possible). The Ethereum team creates motivations for verifiers to introduce the last 100 blocks in accordance with the actual stake ratio by designating a higher number of transaction fees to be redeemed. Any overflow of the fee (due to the staker’s demeanor) will be accumulated to pay the fees later. An obligation exists in each block that comprises data from the last 100 blocks (with a nonce). The accurate top of the chain is the chain with the highest total of load commissions. Over time, the blocks are completed.
In smart contracts, there is the “free option problem,” according to which the recipient (the second or last subscriber) of the proposal of the smart contract is necessary for signing and broadcasting the contract for the purpose of compulsory execution. During this time, the recipient can treat the contract as a free option and reject signing it if the activities are not in their interest. This is aggravated by the fact that smart contracts are most efficient when working with counterparties who can’t be trusted (since this minimizes counterparty risk, and, therefore, data costs).
Plasma Ethereum protocol alone does not solve this problem since there are no guarantees of atomicity with the first and second signature signals for interactive protocols in the blockchain.
With Lightning (including Lightning on top of Plasma), one can make incredibly fast updates with a rational point of localized completeness. Instead of having one payment that gives the option to the last party, the payment can be divided into many small payments. This minimizes the free option to the amount per fraction. Since the second part of the smart contract has only a free option for the amount in fractional form, the value of the free option remains minimal.
In the use cases listed above, it is possible that Lightning can function as the main interface level for fast financial payments/contracts on the Ethereum Plasma platform since Plasma allows updating of registration data with minimal obligations within the main chain.
The Problem of Sharding
With overlaid data sets, there is a significant risk of individual shards rejecting information disclosure. Thus, it would be impossible to produce fraud proofs.
Ethereum aims to solve this using 3 strategies:
At present, work on blockchain communication/Ethereum sharding uses similar methods and goals; for example, the Ethereum Sharding Proposal. This design can be consistent with a higher level. If the root blockchain is sharded, the Plasma chain can work on top of it for wider scalability and other benefits. It can also be a testbed for various Ethereum sharding methods, as there are no consensus shifts needed in Ethereum to begin the main operation.
Off-Blockchain Multiparty Channels
Ethereum aims to create a technique by which users can keep funds in the native main coin of the blockchain without an essential on-chain state. Ethereum Plasma has started erasing the borderline between the on-chain and off-chain.
There are two common problems while trying to set up off-blockchain multi-channels. First, one needs to perform a synchronized state update among all members when an update is required in the system (or else give-and-take the availability of global status updates) and, therefore, should be online. Secondly, supplementing and deleting channel members requires a large blockchain update, listing all members that are added and deleted.
Instead, it would be more desirable to create a mechanism in which many participants can be attached and deleted without considerable parent blockchain state updates, and internal state updates are possible without the involvement of all parties. They would need to participate only if their balances are adjusted or if a Byzantine demeanor revealed.
The general design is a child blockchain that allows holding of the residuals represented in the smart contract on the root blockchain (for instance, Ethereum). The residuals of the smart contract are presented and distributed to the balances of the completed blocks in the child Plasma chain. This allows retention of the native coin in the child blockchain with a complete representation of the remains on the root blockchain, thereby allowing withdrawal of money after the dispute mediation period.
Proof-of-Fraud guarantees that all state transfers are checked and verified. Examples of evidence of fraud are evidence of the effectiveness of transactions (funds obtainable in the current UTXO), Proof-of-State transition (including signature verification for the possibility of the output of results, proof of inclusion/exclusion in blocks, and proof of deposit/withdrawal of funds). Other, more complicated proofs demand an interactive game. The common design will be used to take a functional approach toward block validation. If one creates this consensus mechanism in the Solidity programming language, there will be an extra inlet per function of Merkle proof of block being verified, and the outlet will return whether or not the proof is valid. Then, the consensus confirmation code is simply replicated to process it in the form of a dense proof (so one doesn’t have to process the whole block to create proof of fraud).
However, in order for this design to have minimal proofs, all the blocks have to provide adherence to the Merkleized tree of the current state, tree of the outputs performed, the Merkle transaction tree, and a link to the previously changed state.
Fraud proofs guarantee the security of the system. This assures that a merger of users cannot construct fraudulent blocks avoiding punishments. When a fraudulent block is discovered and verified on the parent blockchain (or the root Plasma chains), the void block is returned back. This motivates separate users to have incentives for Byzantine behavior, which solves the state transition vulnerability problem of the arrangement of predefined functionaries of Bitcoin’s side chains.
As a result, there is greater scalability of state transitions, the capability to block the Plasma blockchain, and assurances that observers who have access to block data can prove an invalid state transition. In other words, payments can happen in this chain with periodic obligations along the parent chain.
Today, distributed registries cannot handle a large number of transactions. One approach to scaling is the separation of various transactions into several block systems. While this allows for a lower transaction demand on each unit, it also means there will be a lower hash power running on every blockchain. In addition, on small side-chains, it’s easy to get enough hash power for attacks. While this offers some degree of scalability, it sacrifices security. The presence of several blockchains also limits transactions between chains of crypto exchanges that charge trade fees, have a long processing time, and are known to be unsafe.
Ethereum has proposed a new and interesting platform to enhance blockchain scalability. The Plasma protocol is still being tested. The Bankex company, which took part in the tests, noted the high throughput: up to 5,000 transactions per second. As a result, an increase in the number of projects on the Ethereum platform will not be associated with network congestion or transaction processing delays. However, we hope for a positive future, which is closely connected with the rapid development of the blockchain-based platforms.