PaperlayerView exact revision on arXiv

Source-linked AI summary

Blockchain for Decentralization of Internet: Prospects, Trends, and Challenges

Javad Zarrin, Phang Hao Wen, Lakshmi Babu Saheer, Bahram Zarrin

arXiv:2011.01096v1cs.NIcs.DCcs.ET

TL;DR

The paper addresses how Blockchain-based methods could support decentralization of the future Internet amid centralized architecture problems. It conducts a literature review of Blockchain-based methods, consensus algorithms, and integrations with future Internet technologies. It identifies PoP, Paxos, and PoAH as more adequate consensus candidates and discusses combinations intended to overcome Blockchain flaws, while noting scalability and performance limitations.

  • Problem

    Centralized Internet architecture creates vulnerabilities and capacity challenges, motivating investigation of Blockchain-based decentralization.

  • Method

    The paper conducts a literature review of Blockchain-based methods, consensus algorithms, and future Internet technology integrations.

  • Results

    The review identifies PoP, Paxos, and PoAH as more adequate consensus algorithms and examines technologies that could optimize Blockchain for Internet decentralization.

  • Takeaways & Limitations

    Blockchain is presented as an effective enabler for a distributed Internet through shared trust across connecting networks.

  • Takeaways & Limitations

    Current Blockchain generations remain slow and unscalable for large transactions, with Smart Contract transmission inefficiency and forking-related delays.

Abstract

from arXiv · show

Blockchain has made an impact on today's technology by revolutionizing the financial industry in its utilization on cryptocurrency and the features it provided on decentralization. With the current trend of pursuing the decentralized Internet, many methods have been proposed to achieve decentralization considering different aspects of the current Internet model ranging from infrastructure and protocols to services and applications. This paper focuses on using Blockchain to provide a robust and secure decentralized computing system. The paper conducts a literature review on Blockchain-based methods capable for the decentralization of the future Internet. To achieve that decentralization, two research aspects of Blockchain have been investigated that are highly relevant in realizing the decentralized Internet. The first aspect is the consensus algorithms, which are vital components for decentralization of Blockchain. We have identified three consensus algorithms being PoP, Paxos, and PoAH to be more adequate for reaching consensus in Blockchain-enabled Internet architecture. The second aspect that we investigated is the impact of future Internet technologies on Blockchain, where their combinations with Blockchain would help to make it overcome its established flaws and be more optimized and applicable for Internet decentralization.

I. INTRODUCTION

The paper examines Blockchain as a means to decentralize the Internet in response to vulnerabilities created by centralized services and infrastructure. It reviews Blockchain-based approaches and their relevance to future Internet architecture.

  • Future Internet motivation: Web 3.0 is presented as a response centered on user autonomy and reduced reliance on centralized services.Its concept places users in responsibility for their data.
  • Centralization challenges: Centralized Internet services create reliability, privacy, security, scalability, and availability problems for users and providers.The paper notes single points of failure, exposed user data, limited resources, and regional or time-specific shortages.
  • Centralization challenges: Large Internet corporations and service providers can maintain, control, and administrate user data, access, and activities.The paper associates this trusted-party role with possible surveillance, censorship, and abuse of trust.
  • Paper scope: The paper provides a systematic review of Blockchain-based solutions for Internet decentralization across Internet components.The review focuses on Blockchain rather than other decentralization approaches and excludes Blockchain for IoT security because of space limitations.

II. UNDERSTANDING THE CONTEMPORARY INTERNET ARCHITECTURE

The contemporary Internet relies on centralized routing, services, and protocols across a large multi-layer architecture. This centralization creates control, security, capacity, addressing, and interoperability challenges.

  • Protocol challenges: Application-layer protocols such as HTTP and HTTPS rely on the centralized client-server model, requiring a communication method to interoperate with Blockchain.The paper also distinguishes HTTPS handshakes from Blockchain cross-referencing security measures.
  • Protocol challenges: Transport-layer weaknesses include enumeration attacks, fingerprinting, and SYN floods, motivating consideration of alternative security solutions.These attacks can extract system information, identify open ports, or overwhelm systems.
  • Addressing challenges: IPv4 address-space exhaustion motivates IPv6, which expands addresses from 32 bits to 128 bits and addresses several IPv4 limitations.The paper states that complete transition from IPv4 to IPv6 is impossible.
  • Centralized architecture: Centralized Internet architecture routes users through singular administrative points such as ISPs and retains centralized elements in DNS governance and operation.The paper identifies ISP networks as centrally administered access points and notes remaining DNS centralization.
  • Challenges of Centralized Internet: Centralized networks expose users to single points of failure, single points of control, compromised entry points, and growing storage requirements.These risks affect infrastructure, databases, and the capacity needed for expanding Internet information data.

C. Types of Decentralization

The paper distinguishes fully decentralized and distributed networks as two approaches to Internet decentralization. It presents the distributed approach as more plausible for transitioning existing Internet systems.

  • Decentralized Internet examples: Decentralized Internet projects use P2P technology for shared data layers and emphasize anonymous access, data security, and cooperation.Examples include TOR, Zeronet, and I2P.
  • Types of Decentralization: A fully decentralized network spreads trust and control across anonymous users rather than a centralized point.Participating computers can follow their own configurations instead of a dictated standardization.
  • Types of Decentralization: A distributed network interconnects participating computers and allows centralized legacy systems to operate within a decentralized network.The paper considers this method more plausible for a complete Internet transition.
  • Blockchain as an enabler: Blockchain is presented as a prominent option because it shares traits with decentralized projects, including P2P communication, data storage, encryption, and shared trust.The paper connects these capabilities to Blockchain components illustrated in its technology overview.

III. UNDERSTANDING BLOCKCHAIN-BASED DECENTRALIZATION

The paper describes Blockchain as a decentralized database whose nodes validate and append transactions while preserving shared, immutable records. Its focus is decentralization, with other Blockchain characteristics outside scope.

  • Blockchain process: A Blockchain transaction is requested by a node, packed into a block, broadcast for validation, and appended after successful verification.The process ends with the verified block stored at the end of the Blockchain.
  • Blockchain characteristics: Blockchain decentralization avoids third-party validation and gives nodes equal voting rights used by consensus algorithms.The paper identifies consensus as the mechanism that dictates the Blockchain.
  • Paper scope: The paper focuses on decentralization and excludes persistency, anonymity, and auditability from its scope.These characteristics are described but not examined as primary research topics.
  • Blockchain fundamentals: Blockchain stores decentralized-network data in a distributed database whose append-only ledgers record timestamped transactions across nodes.This structure tracks transactions without requiring a central verification figure.
  • Blockchain characteristics: Immutable storage keeps database records unalterable across nodes and supports ledger integrity through encryption.Each database is retained in every node with a reference to its immutable Blockchain history.

3) Consensus Algorithm:

Blockchain consensus algorithms govern how nodes agree to update the ledger, while Blockchain generations and designs address decentralization, scalability, performance, privacy, and mining concerns.

  • Consensus Algorithm: Consensus algorithms determine how nodes reach agreement and append ledger modifications as new blocks.They moderate the Blockchain by specifying how nodes agree and update the network.
  • Types of Blockchains: Public Blockchains permit anyone to verify and participate in consensus, whereas consortium and private Blockchains restrict these roles to selected nodes.Public systems are permissionless; consortium and private systems are permissioned.
  • Generations of Blockchain: Third-generation Blockchain proposals emphasize interoperable structures and protocols, with applications targeting higher transaction scalability and efficiency.These developments motivate using Blockchain as a decentralized software architecture for the future Internet.
  • Limitations of Blockchain: Blockchain remains only partially decentralized because network pressures can form centralized nodes.This creates limitations for current second-generation Blockchains.
  • Limitations of Blockchain: Scalability bottlenecks arise from limited block capacity, verification requirements, and increased storage demands, though parallel mining and edge or fog computing are proposed remedies.Chu-ko-nu Mining uses asynchronous consensus zones and reports over a thousand times the throughput and two thousand times the capacity of Bitcoin and Ethereum.
  • Limitations of Blockchain: Current Blockchain limitations also include slow performance, privacy vulnerabilities, and selfish mining that wastes resources or creates shorter private branches.Forking can cause network delays exceeding 1000 seconds, while keys may expose private information and transaction identities may be traced.

IV. INVESTIGATING THE CONSENSUS ALGORITHM IN BLOCKCHAIN FOR DECENTRALIZATION

The paper reviews consensus algorithms for Internet decentralization, emphasizing that suitable designs should avoid trusted third parties while supporting dependable decentralized operation. It identifies PoP, Paxos, and PoAH as candidates, while noting important trade-offs in existing protocols such as PoW and PoET.

  • Consensus Algorithm Investigation: A decentralized Blockchain consensus algorithm should avoid trusted third-party services, making permissioned Blockchains the favored choice for dependable decentralized operation.Fog and edge architectures are also relevant to Blockchain deployment for IoT.
  • Consensus Algorithm Investigation: The review compares consensus protocols for Internet compatibility, but recommendations remain provisional because the algorithms have varying uncertainty.The reviewed algorithms are presented in Tables III and IV.
  • Proof-Based Consensus: Proof-based consensus lets competing nodes solve cryptographic problems, with the successful node earning the right to append the next block.This approach is widely used in permissionless Blockchains.
  • Proof-of-Work: PoW rewards computational competition but causes substantial power wastage, limiting its suitability for decentralized Internet applications.Its consensus cycle repeatedly uses computational power to solve mathematical puzzles.
  • Proof-Of-Elapsed Time (PoET): PoET lowers power consumption and offers low latency and high throughput, but dependence on Intel SGX introduces a centralized point.The protocol is therefore presented as promising for resource-constrained IoT devices while conflicting with decentralization goals.

3) Proof-Of-Search (PoS):

The reviewed consensus approaches trade computational efficiency, IoT suitability, scalability, and decentralization against implementation or applicability limitations.

  • 3) Proof-Of-Search (PoS):: PoS provides computational service within grid-computing infrastructures suited to large networks such as data centers.Its optimization process requires each node to check many possible solutions, creating large computation requirements that may hinder IoT performance and compatibility.
  • 4) Proof-Of-Authentication (PoAh):: PoAh targets IoT through lightweight block verification that authenticates both the block and its source.Completed verified transactions generate trust values, while efficient verification supports integration with fog computing and edge infrastructure.
  • 5) Proof-Of-Property (PoP):: PoP reduces transaction information, storage, and processing requirements through address-tied proofs that store Blockchain state in each new block.The protocol is described as lightweight and scalable, with potential IoT suitability, but it has not yet been successfully applied in industry and remains immature.
  • Consensus algorithm comparison: Consensus algorithms unsuitable for Internet architecture may depend on cryptocurrency-specific data or activity-based node features that complicate transactions.The review also describes Byzantine-based consensus as addressing false messages from failed nodes during agreement.
  • 1) Practical Byzantine Fault Tolerance (PBFT):: PBFT offers high throughput, low latency, and lower power usage than PoW, but its requirement for all nodes to participate limits permissionless-Blockchain scalability.PBFT reaches consensus with agreement from two-thirds of the nodes but has high network overhead and low exploit tolerance.

2) Delegated Byzantine Fault Tolerance (dBFT):

The reviewed BFT, crash-based, and Hyperledger approaches provide varied performance and scalability characteristics, but several remain unsuitable for decentralized Internet consensus under specific constraints.

  • 2) Delegated Byzantine Fault Tolerance (dBFT):: dBFT improves scalability over PBFT by allowing selected representative nodes to participate instead of requiring every node.Its average 15-second block-creation latency is reported as unacceptable, making dBFT unsuitable as a consensus candidate for the Blockchain.
  • 3) Stellar Consensus Protocol (SCP):: SCP forms quorums from trust-based quorum slices in a peer-to-peer web-like structure and can provide high throughput with low power usage.Significant network overhead causes latency, while selecting an incorrect quorum slice creates a security issue; these limitations make SCP unsuitable for consensus.
  • Related alternatives: Ripple reduces Blockchain latency and tolerates up to 20% faulty nodes, but its monetary focus and speed do not meet IoT requirements.
  • 4) Hyperledgers:: Hyperledger projects primarily target permissioned Blockchains, but Fabric depends on certification from a centralized point through Chaincode.This dependence makes Hyperledger Fabric less ideal for decentralization, while Sawtooth remains insufficiently developed for consideration.
  • 4) Hyperledgers:: Hyperledger Burrow may halt without specific Blockchain roles, and Hyperledger projects require a leader in permissioned consensus; Iroha offers potential through its mobile design.
  • Proof-based consensus: PoA restricts block creation to a fixed Byzantine-selected node set, making it suited to enclosed networks with an administrator rather than open decentralization.
  • Crash-based consensus: Crash-based consensus accounts for crashed nodes during agreement but cannot sustain complete crash tolerance for the Blockchain system.

1) Paxos:

Paxos offers strong safety and crash tolerance but is difficult to understand, designed for small enclosed networks, and centralized-like in operation. The paper nevertheless considers it a plausible Internet consensus candidate requiring adaptation, alongside technologies that could reduce Blockchain resource demands.

  • 1) Paxos:: Paxos is a theoretical consensus algorithm with crash tolerance of up to 50%, but it is challenging to understand and implement.Its design includes leader and follower roles.
  • 1) Paxos:: Paxos was designed for smaller enclosed networks, making it unsuitable for direct Internet implementation despite useful ballot and anchoring safety features.
  • 1) Paxos:: A dominating leader role makes Paxos operate in a centralized-like way despite the possibility of distributed implementation.Followers acknowledge the leader and provide votes, while the leader makes progress within the protocol.
  • 1) Paxos:: PoP, Paxos, and PoAh are identified as consensus candidates, with Paxos considered plausible but requiring development to adapt it for Blockchain.PoP reduces storage and processing needs, while PoAh is described as robust, scalable, and secure for Internet-related infrastructure.
  • 1) Paxos:: Cloud computing can increase Blockchain security and scalability while reducing transaction data storage and network overhead.The paper also describes IoT and cloud computing as improving verification efficiency and enabling more consensus algorithms.
  • 1) Paxos:: Blockchain, cloud computing, and IoT are projected to converge into Blockchain-of-Things, but standardized communication protocols remain absent.Protocol development is identified as a priority before integration.

C. Graphchain

Graphchain replaces Blockchain’s linear chain with a decentralized graph intended to improve scalability and transaction performance. Its benefits are accompanied by a centralization issue from shared common descendants, while edge and fog computing provide complementary resource and processing capabilities.

  • C. Graphchain: Graphchain replaces Blockchain’s linear network structure with a graph data structure.It retains Blockchain components while using a decentralized graph of self-scaling and self-regulated cross-verifying transactions.
  • C. Graphchain: Graphchain distributes transaction data among data shards and uses parallel mining to improve scalability and transaction-processing performance.
  • C. Graphchain: Graphchain can incorporate semantic technology to add relations and meaning to its distributed-ledger data structure.
  • C. Graphchain: Graphchain develops centralization through a common descendant shared by newly created transactions, despite its decentralized design objective.The paper presents this drawback alongside Graphchain’s scalability and performance benefits.
  • C. Graphchain: Edge computing expands cloud computing by distributing resources toward the network edge, where Blockchain could use edge processing for connected nodes.Edge and fog computing are described as offering scalability, security, and performance benefits.
  • C. Graphchain: Fog computing distributes computing, control, storage, and networking resources across the Cloud-to-Edge continuum, reducing storage needs for IoT and Blockchain communication.Blockchain-based fog clusters use consensus algorithms to coordinate network computers.

G. Data Networking

The data-networking discussion examines standardization, software-defined control, and information-centric approaches for handling Internet-scale data. It describes ICN components including named objects, naming and security, APIs, and caching, while noting single-point-of-failure and routing considerations.

  • G. Data Networking: Network configuration standardization is necessary because conflicting protocols can hinder performance, increase costs, and reduce scalability and reliability.The paper notes that standardization may require authoritative management despite decentralization goals.
  • G. Data Networking: SDN removes device control functionality, uses OpenFlow, relies on an external controller, and supports programmable applications above the data plane.
  • G. Data Networking: Information-Centric Networking centers on content data and offers scalable, cost-efficient handling of expanding IP traffic through persistence and unique naming.
  • G. Data Networking: Named Data Objects combine a unique identifier, data, and metadata, using hierarchical or self-certifying naming schemes.Self-certifying names embed a hash containing prefixes into the data.
  • G. Data Networking: ICN’s naming and security component includes external name resolution, which can create a single point of failure, or direct routing from requester to data source.
  • G. Data Networking: ICN uses APIs to request and deliver named data objects, while caching lets nodes satisfy requests using stored copies.Simulations indicate edge caching can improve IoT data distribution efficiency.

3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):

CCN and NDN replace address-centered communication with named-data networking to improve content routing and accommodate emerging demands such as Big Data. The broader Blockchain and future-technology discussion emphasizes integration opportunities alongside standardization and centralization trade-offs.

  • 3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):: Content-Centric Networking makes content nameable and routable, communicating through named data rather than TCP/IP addresses.
  • 3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):: NDN evolves CCN by using named data to address emerging demands such as Big Data that the paper associates with TCP/IP obsolescence.
  • 3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):: NDN reshapes the TCP/IP hourglass by replacing IP with named Content Chunks.The paper presents this architecture as combining networking, storage expansion, and processing capabilities.
  • 3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):: The paper relates NDN’s data naming to Semantic technology through links, relevancy, and meaning assigned to Internet data chunks.
  • 3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):: Blockchain integration with future technologies is framed as an opportunity to improve Blockchain, but adoption requires decisions about technology trade-offs and standardization.
  • 3) Content-Centric-Networking (CCN) & Named-Data Networking (NDN):: Graphchain may optimize Blockchain while introducing route centralization, and cloud middleware standardization may reduce performance and scalability.The paper presents balancing these trade-offs as a challenge for future Internet architecture.

C. Development Trends

The section traces decentralization challenges and development trends involving IoT, decentralized infrastructure, and Blockchain data structures. It also highlights unresolved trade-offs, cryptographic alternatives, and consensus technologies for future Internet architectures.

  • IoT and future Internet: IoT is driving new Internet architectures in which smart devices interact across connected networks.The paper considers how IoT should interact with proposed Blockchain Internet architectures.
  • Decentralization trade-offs: Web 2.0 centralized functions that had made the Web more optimized, raising whether comparable efficiency can be achieved after decentralization.The migration from decentralized Web 1.0 to Web 2.0 introduced centralized services and greater functionality.
  • Decentralization trade-offs: Personalized applications and advertisements create a quality-of-service trade-off because complete decentralization may remove centralized profiling of user data.The paper frames acceptance of this loss as an open design question during migration toward decentralization.
  • Emerging applications: Blockchain-based decentralized infrastructure is discussed alongside IoT, Graphchain, and NDN for possible future battlefield networks using real-time information.The paper presents military networking as a possible future application of decentralized infrastructure.
  • Blockchain infrastructure: Merkle trees optimize distributed ledger storage and verification, while their use raises the need to consider alternative or more optimized Blockchain encryption methods.Merkle roots verify ledger integrity after repeated hashing of multiple information items.
  • Consensus and enabling technologies: The study identifies consensus algorithms and relevant technologies as the two Blockchain research aspects most important to realizing Internet decentralization.It highlights Proof-Of-Property, Paxos, and Proof-Of-Authority as options, while noting that future technologies may replace proposed ones.
Loading 2011.01096v1…