Scalability: ICP's Balanced Approach to Scalability and Cost-effective

ICP leverages chain key technology, high-performance node configuration, optimized consensus mechanism, and chain fusion to enhance scalability.

Key Takeaways

• Web3 envisions a decentralized Web that promises transparency, security, and enhanced user autonomy. However, Key challenges of mass adoption of Web3 include: scalability, user experience, security, interoperability, regulation and compliance, adoption barriers. This article mainly focuses on scalability issues.

• Mainstream solutions for improving blockchain scalability include Layer 1 solutions, Layer 2 solutions, interoperability, and data storage optimization. Bitcoin and Ethereum have implemented various solutions, including as SegWit and the Lightning Network for Bitcoin, and Layer 2 solutions (such as rollups) for Ethereum, to solve throughput and latency issues.

• Achieving a Web3 environment involves integrating various technologies and tools to build decentralized applications (dApps) and services. In this article, we selected the Internet Computer Protocol (Hereinafter referred to as “ICP”) as the research object to discuss its contribution to scalability improvement considering the project’s key features: chain key technology, optimized consensus mechanisms, high-performance node configuration, and chain fusion technology.

Background

Since the debut of Bitcoin in 2009, which ignited the cryptocurrency's innovation trigger, developers in the space have been busily trying to shorten the slope of the enlightenment stage and hasten the onset of the productivity plateau.

Gavin Wood, a co-founder of Ethereum, first used the term "Web3" in 2014. It refers to a network environment that emphasizes security, privacy, and user autonomy by combining decentralization, blockchain technology, smart contracts, and cryptocurrencies. Due to a rise in interest in cryptocurrencies from investors, large tech businesses, and enthusiasts, the concept became one of the main stories surrounding the bull market fervor in 2021. But as the market shrank, we saw that there are obstacles in the path of achieving broad Web 3 acceptance. Although Web 3 has not been completely refuted, there are a lot of obstacles and doubts around it. Although the idea of a decentralized network is intriguing, a number of problems must be resolved before it can be widely embraced and shown to be sustainable in the long run.

Key Challenges

• Scalability: The mainstream adoption of blockchain technology is hindered by its current networks' inability to scale, which results in expensive prices and sluggish transaction rates.

• User Experience: When compared to conventional Web2 services, the user experience in Web3 apps is frequently more sophisticated and less intuitive, which makes it challenging for regular users to interact.

• Security Concerns: Web3 applications have been the target of scams and cyberattacks despite the promise of improved security, raising questions about the security of decentralized platforms.

• Interoperability: The lack of interoperability between different blockchain networks creates fragmentation and limits the seamless use of Web3 services across platforms.

• Regulation and Compliance: The decentralized nature of Web3 poses challenges in terms of regulatory oversight and compliance, which could lead to legal uncertainties and potential crackdowns.

• Adoption Barriers: The transition from centralized to decentralized systems requires significant shifts in infrastructure, mindset, and incentives, which can be slow and resistant to change.

In this article, we'll discuss the various approaches taken by the industry to address the scalability issue in brief and concentrate on the benefits and drawbacks of using ICP as the Web3 infrastructure.

Mainstream Solutions to Improve Scalability

Blockchain networks such as Bitcoin and Ethereum are unable to effectively handle large numbers of transactions due to scalability problems, which results in network congestion, sluggish transaction rates, and expensive fees. The following are some common ways to improve the scalability of blockchains:

• Layer 1 Solutions: Consensus Mechanism Optimization and Sharding.

• Layer 2 Solutions: State Channels, Sidechains and Rollups.

• Interoperability: Cross-Chain Communication Protocols.

• Data Storage Optimization: Off-Chain Data Storage and Zero-Knowledge Proofs (ZK-Proofs).

The introduction of Segregated Witness (SegWit) has been a positive development for Bitcoin. By isolating transaction signatures from transaction data, SegWit expands the effective block size. This modification relieves some of the congestion by allowing more transactions to fit in each block.

Meanwhile, Off-chain transactions are also made possible by Lightning Network technology, which tackles scalability issues and lowers transaction latency. The amount of labor on the main network can be greatly decreased by users by creating payment channels where transactions are cheap and instantaneous, and only the final state is recorded on the blockchain.

Throughput and latency are predicted to improve for Ethereum with the adoption of Proof of Stake (PoS) and layer 2 solutions like zk-Rollups and Optimistic Rollups. By processing transactions off-chain and sending a summary of those transactions to the main chain on a regular basis, these layer 2 solutions sigthroughput, cut down on latency, and maximize the use of scarce block space. Furthermore, the network will be split up into smaller segments, or shards, by Ethereum 2.0's sharding mechanism. Each shard will be able to handle its own smart contracts and transactions. By doing this, the network's capacity will rise dramatically and less off-chain data storage will be required.

Beyond Bitcoin and Ethereum, various blockchain projects have worked on a variety of novel techniques to improving scalability. This article investigates scalability augmentation approaches, using the Internet Computer Protocol as an example.

During the scaling solution research, we discovered the Internet Computer Protocol, and the project's strong TPS performance piqued our interest in conducting additional research on it.

The following table contains statistics on the performance and cost of the main blockchain Networks VS conventional payment systems.

ICP's Balanced Approach to Scalability and Cost-effective

ICP intends to build a decentralized and scalable global internet infrastructure in which smart contracts, known as "canisters," can operate at the same speed, security, and scale as standard online apps.

ICP envisions a fully decentralized internet where services, platforms, and data are no longer controlled by centralized entities but are governed and run on a network of independent data centers.

Scalability is essential to ICP's concept, as it is to other blockchain networks, because a decentralized internet must be able to manage a high number of data and transactions in order to be competitive and sustainable with old centralized systems. A network with high scalability can allow widespread adoption, expand, and run at web speed without compromising decentralization or security.

ICP uses a combination of chain key technology, high-performance node configuration, optimized consensus mechanism, and chain fusion to increase its scalability and target to realize its goal of creating a decentralized internet that can serve massive applications.

Consensus Mechanism Optimization

Reaching consensus across tens of thousands of nodes is the most time-consuming procedure for Bitcoin and Ethereum that affects TPS the most. In particular:

• Broadcasting Transactions and Blocks: A transaction that a user starts needs to be broadcast to every network node. It takes time for each node to receive, verify, and rebroadcast the transaction—especially in big, decentralized networks.In a same vein, a newly mined block needs to spread around the network. The block must be downloaded, checked, and added to each node's local copy of the blockchain.

• Verification and Validation: Every node independently confirms that blocks and transactions adhere to the network's guidelines. This includes verifying signatures and examining double-spending.

• Consensus Mechanism: Achieving consensus on the subsequent block to be added to the blockchain.

In PoW systems like Bitcoin, miners compete to solve a cryptographic puzzle, which is computationally intensive and time-consuming. The first miner to solve the puzzle broadcasts the new block to the network. 

In PoS systems like Ethereum 2.0, validators propose and vote on the next block. This involves multiple rounds of communication and voting, which can be time-consuming.

Therefore, according to the consensus mechanisms described above, nodes need to communicate more often to obtain consensus as the network develops, which raises overhead and lowers TPS.

Next, we will describe the optimization that IC made to the consensus mechanism, as outlined in criteria below.

• BLS Signature Aggregation: Combines multiple signatures into one, reducing verification load.

• Simplified Consensus Process: Streamlines node communication, speeding up consensus.

• Threshold Relay Mechanism: Quick consensus with a threshold number of node approvals.

• Random Beacon: Randomly selects the next block proposer for fast, unbiased generation.

By using a streamlined consensus procedure with BLS signature aggregation, ICP successfully lowers this burden. Multiple signatures can be combined into a single one using BLS signatures. Nodes can combine several signatures into a single one during transaction verification, necessitating just one verification. As a result, the bandwidth and processing power required for signature verification are greatly decreased.

If 1,000 transactions are involved, each requiring a signature verification, and the verification time is 10 milliseconds for each signature, the overall verification time would be 10 seconds. Efficiency is greatly increased when 1,000 signatures are combined into one using BLS signature aggregation, which takes only 10 milliseconds.

Furthermore, through the threshold relay mechanism, only the threshold number of nodes (e.g., two-thirds of the total nodes) need to sign a block for it to be considered valid. This mechanism reduces the number of communications between nodes, thus decreasing network latency.

Now Let’s assume a network with 100 nodes, the previous consensus mechanism might require each node to communicate with the other 100 nodes, whereas the threshold relay only requires communication with approximately 67 nodes, thereby reducing communication rounds.

ICP uses fast block generation to greatly enhance TPS. "Fast block generation"—what is that? A random beacon in conjunction with a threshold relay allows for fast block generation.

The next block proposer is chosen by the random beacon, which uses BLS signatures to generate an unpredictable and unmanageable random number. By preventing nodes from knowing ahead of time if they will be chosen, this random selection reduces the likelihood of attacks and guarantees fairness. Wait times can be significantly decreased by the random beacon selecting the proposer in a matter of milliseconds, provided that each block generating round takes one second.

Once the required number of signatures is gathered, nodes can swiftly come to a consensus and produce a block thanks to the threshold relay mechanism. If the threshold is set at, say, 67%, a block can be generated right away without waiting for consensus from every node. Block generation speed can be increased by the threshold relay completing consensus in a few seconds if it takes 10 seconds for all nodes to reach consensus.

Chain Key Technology

ICP network uses sharding to achieve scalability by dividing the network into multiple subnets, each functioning as an independent blockchain. These subnets handle their own transactions and host canisters (smart contracts). As demand grows, new subnets can be dynamically created, distributing the workload and enabling the network to scale horizontally without a predefined limit. Each subnet operates in parallel, allowing the network to expand and manage increasing demand effectively. Image a social media platform with millions of users built on ICP network. As more users join, additional subnets are created to handle the increased data and processing needs. The platform continues to operate smoothly, regardless of the number of users.

Chain Key Technology is a cryptographic innovation that allows subnets within the ICP to communicate with each other securely and efficiently. Each subnet maintains a chain key, a single public key representing the entire subnet. This key enables subnets to verify and interact with each other without requiring the entire transaction history, reducing the communication overhead. It enables seamless cross-subnet transactions and data exchanges. For instance, when a user interacts with a canister on one subnet that needs data from another subnet, Chain Key Technology allows this interaction to happen instantly and securely, without waiting for multiple confirmations.

Canisters are not tied to a single subnet and can be moved between subnets as needed. When a subnet becomes too congested, canisters can be migrated to less busy subnets without interrupting their operations. This ensures that the network remains balanced and prevents any single subnet from becoming a bottleneck, and allows the network to balance the load across subnets and optimize resource utilization.

Node Configuration

High-performance node devices can also greatly increase network TPS, in addition to network sharding. ICP nodes are made for high-performance networking and computing applications and are of server-grade quality. They have large RAM (16 x 32GB RDIMM), dual-port 10G network connections, high-speed NVMe storage, and dual-socket AMD EPYC Milan CPUs. These specs provide a strong, dependable, and scalable infrastructure required for high transaction processing and data storage demands, surpassing the capabilities of consumer-grade hardware.

Node Machine Spec Comparison

Additionally, the network's high bandwidth and low latency capabilities are crucial. ICP nodes are typically located in data centers with high-speed internet connections, which ensure rapid data transmission and reception. Low latency is particularly vital for the consensus process, as it reduces the time required for nodes to communicate and agree on the state of the blockchain, thereby maintaining the network's integrity and speed.

Chain Fusion

The purpose of ICP Chain Fusion is to facilitate smooth communication and exchange of data, transactions, and smart contract features between several blockchains. By uniting many blockchain networks into a single system, the mechanism seeks to redefine Web 3, which is essential for resolving scalability problems that other blockchains face.

The contribution of ICP Chain Fusion to scalability benefits, we tried to summarize into items below:

• Horizontal Scaling Without Performance Loss
  Chain Fusion's dynamic integration of subnet blockchains ensures horizontal scaling, which allows the network to handle a larger volume of transactions and support more complex applications without suffering from performance degradation.

• Decentralized Structure with Retained Security
  Unlike traditional systems that rely on centralized servers or external databases, Chain Fusion maintains a decentralized structure. Each integrated blockchain retains its own consensus mechanisms and security protocols, preserving decentralization and resistance to central points of failure.

• Optimized Resource Utilization
  The network optimizes resource utilization by allocating computationally intensive tasks to blockchains with high processing power, while those with efficient data storage handle storage-heavy tasks. This ensures the network operates at peak efficiency, avoiding bottlenecks and inefficiencies common in traditional systems.

• Cross-Chain Synergy
  The integration of networks like BTC, ETH, and ICP allows for enhanced transaction throughput, improved security, and a unified smart contract and dApp ecosystem. Especially with ICP's canister smart contracts, the synergy of these networks supports the development of more powerful and scalable decentralized applications (dApps).

High-speed Blockchain Centralization Risk

Since we've already covered the benefits that ICs offer in terms of scaling, are there any concerns involved? Yes, is the response.

The degree of decentralization of the blockchain network could theoretically be impacted by the deployment of server-level node hardware. It raises the bar for individual participation due to the high cost and complex operating requirements. A Gen 2 node of the ICP based on geography can be acquired and maintained for a total cost ranging from 31000 to 57500 in XDR. XDR are additional foreign exchange reserve assets that are defined and managed by IMF.

Therefore, how to balance the degree of decentralization, network size, and cost is a significant concern for ICP. 

Let us now return to discussing the countermeasures that ICP is currently implementing to address this risk.

Deterministic Decentralization in Nodes Distribution

In blockchain networks, increasing the number of nodes enhances decentralization but also raises costs for node providers. Total decentralization is challenging, even Bitcoin and Ethereum face centralization risks from mining pools and staking whales. A practical solution is "decentralization with multiple centers," where control is distributed among multiple independent entities.

What is Deterministic decentralization? In simpler terms, it means that the process of distributing control or power is done according to a set of predefined rules or algorithms, making the outcomes both reliable and transparent. ICP uses this mechanism by setting a target topology to achieve optimal decentralization across nodes (considering jurisdictions, geography and ownership, etc.). The target topology focusing on four distinct metrics: node provider, data center, data center provider, and country.

Currently, ICP has 1,493 nodes across 37 subnets, with 559 active machines. The nodes on the IC exhibit significant diversification for data center,data center provider, and node provider. Each node in a subnet is operated by a different node provider, data center, and data center provider, with geographical restrictions ensuring that no more than two or three nodes are in the same country. 

 ICP has been working on decentralized node distribution in national geolocation for a while now, mostly in the US and Switzerland.

Network Nervous System (NNS) Contributes to Decentralization

ICP's Network Nervous framework (NNS) is a creative, automated governing framework that makes decentralization easier. By empowering ICP token holders to participate in governance, NNS ensures that decision-making is decentralized and community-driven. Token holders can vote on proposals for network upgrades and changes, effectively steering the network's development. Furthermore, NNS dynamically manages nodes and subnets, distributing them across various independent operators to avoid centralization. This setup allows for seamless protocol upgrades without the need for hard forks, maintaining continuous network operation and improvement. By spreading control and fostering community participation, NNS promotes a high degree of decentralization while ensuring the network remains efficient and secure.

Source:Dashboard.internetcomputer.org

Conclusion

ICP does demonstrate a great deal of promise for scalability improvement; nevertheless, market testing of industry adoption has not yet occurred. We will keep an eye on and research further. We also invite ICP researchers to contribute insightful ideas and collaborate on brainstorming sessions.

For Web3 to succeed, scalability is essential because it dictates whether decentralized networks can match centralized systems in terms of effectiveness, speed, and user experience. It is still very difficult to achieve scalability without sacrificing interoperability, security, or decentralization. To get over these challenges and fulfill the promise of Web3, more innovation in consensus algorithms, node performance, and cross-chain connectivity is required. The future of Web3 will depend on a well-balanced strategy that improves scalability while upholding the fundamental ideas of security and decentralization as the industry develops.

Reference

• DFINITY Foundation. (2021). Internet Computer White Paper. https://internetcomputer.org

• Nakamoto, S. (2016). Segregated Witness (SegWit) Implementation. Bitcoin Improvement Proposal (BIP) 141. https://github.com/bitcoin/bips/blob/master/bip-0141.mediawiki

• Dodo Khan.(2021).Systematic Literature Review of Challenges in Blockchain Scalability. Appl. Sci. 2021, 11(20), 9372; https://doi.org/10.3390/app11209372

• Lightning Labs. (2018). The Lightning Network: Scalable Off-Chain Instant Payments. https://lightning.network/lightning-network-paper.pdf

• ICP Developer Forum,”IC topology series - Node diversification”, available at forum.dfinity.org.