Blockchain Scalability

Preface
Acknowledgments
Contents
1 Blockchain Scalability Fundamentals
	1.1 Overview
	1.2 Preliminaries of Blockchains
		1.2.1 Prime Blockchain Platforms
			1.2.1.1 Bitcoin
			1.2.1.2 Ethereum
			1.2.1.3 Hyperledger Fabric
			1.2.1.4 EOSIO
		1.2.2 Consensus Mechanism
		1.2.3 Scalability of Blockchains
			1.2.3.1 Off-Chain Techniques
			1.2.3.2 DAG
			1.2.3.3 Sharding Technique
			1.2.3.4 Cross-Shard Transactions
	1.3 Theories to Improving the Performance of Blockchains
		1.3.1 Latest Theories to Improving Blockchain Performance
			1.3.1.1 Throughput and Latency
			1.3.1.2 Storage Efficiency
			1.3.1.3 Reliability of Blockchains
		1.3.2 Scalability-Improving Solutions
			1.3.2.1 Solutions to Sharding Blockchains
			1.3.2.2 Multiple-Chain and Cross-Chain: Interoperability Amongst Multiple Blockchains
		1.3.3 New Protocols and Infrastructures
			1.3.3.1 New Protocols for Blockchains
			1.3.3.2 New Infrastructures and Architectures for Blockchains
	1.4 Various Modelings and Techniques for Better Understanding Blockchains
		1.4.1 Graph-Based Theories
		1.4.2 Stochastic Modelings
		1.4.3 Queueing Theories for Blockchain Systems
		1.4.4 Analytical Models for Blockchain Networks
		1.4.5 Data Analytics for Cryptocurrency Blockchains
			1.4.5.1 Market Risks Detection
			1.4.5.2 Ponzi Schemes Detection
			1.4.5.3 Money-Laundering Detection
			1.4.5.4 Portrait of Cryptoeconomic Systems
	1.5 Useful Measurements, Datasets and Experiment Tools for Blockchains
		1.5.1 Performance Measurements and Datasets for Blockchains
		1.5.2 Useful Evaluation Tools for Blockchains
	1.6 Open Issues and Future Directions
		1.6.1 Performance-Improving Issues
			1.6.1.1 Scalability Issues
			1.6.1.2  Resilient Mechanisms for Sharding Technique
			1.6.1.3 Cross-Shard Performance
			1.6.1.4 Cross-Chain Transaction Accelerating Mechanisms
			1.6.1.5 Ordering Blocks for Multiple-Chain Protocols
			1.6.1.6 Hardware-Assisted Accelerating Solutions for Blockchain Networks
			1.6.1.7 Performance Optimization in Different Blockchain Network Layers
			1.6.1.8 Blockchain-Assisted BigData Networks
		1.6.2 Issues for Better Understanding Blockchains Further
		1.6.3 Security Issues of Blockchains
			1.6.3.1 Privacy-Preserving for Blockchains
			1.6.3.2 Anti-cryptojacking Mechanisms for Malicious Miners
			1.6.3.3 Security Issues of Cryptocurrency Blockchains
		1.6.4 Powerful Experimental Platforms for Blockchains
	1.7 Conclusion
	References
2 Overview to Blockchain Scalability Challenges and Solutions
	2.1 Overview
	2.2 Scalability Issue of Blockchain
	2.3 Taxonomy of the Approaches to Solving the Scalability of Blockchain
		2.3.1 Layer1: On-Chain Solutions
			2.3.1.1 Solutions Related to Block Data
			2.3.1.2 Different Consensus Strategies
			2.3.1.3 Sharding
			2.3.1.4 DAG (Directed Acyclic Graph)
		2.3.2 Layer2: Non On-Chain Solutions
			2.3.2.1 Payment Channel
			2.3.2.2 Sidechain
			2.3.2.3 Off-Chain Computation
			2.3.2.4 Cross-Chain Techniques
	2.4 Future Directions and Open Issues
		2.4.1 Layer-1
			2.4.1.1 Block Data
			2.4.1.2 Sharding Techniques
		2.4.2 Layer-2
		2.4.3 Layer-0
	2.5 Conclusion
	References
3 On-Chain and Off-Chain Scalability Techniques
	3.1 Overview
	3.2 Related Work
	3.3 Two-Layer Scaling Sharing Framework Based on Large-Scale Wireless Networks
		3.3.1 Framework Overview
		3.3.2 Two-Layer Scaling Protocol
		3.3.3 Automated Sharing Transaction Workflows
		3.3.4 Advantages of Framework
			3.3.4.1 Real-Time
			3.3.4.2 Trusted Data Interaction
			3.3.4.3 Fine-Grained Transaction Support
	3.4 Case Study: ITS-Data-Sharing Economy
		3.4.1 Proof of Concept Implementation
		3.4.2 Performance Evaluation
	3.5 Conclusion and Open Issues
	References
4 Layered Sharding on Open Blockchain
	4.1 Overview
	4.2 System and Threat Model
		4.2.1 System Model
		4.2.2 Threat Model
	4.3 System Design
		4.3.1 Layered Sharding Formation
		4.3.2 Cross-Shard Block Design
		4.3.3 Layered Sharding Consensus
		4.3.4 Design Refinements
	4.4 Analysis
		4.4.1 Security Analysis
		4.4.2 Scalability Analysis
		4.4.3 Performance Analysis
	4.5 Evaluation
		4.5.1 Implementation
		4.5.2 Setup
		4.5.3 Throughput
		4.5.4 Confirmation Latency
		4.5.5 Storage Overhead
		4.5.6 Commit Ratio
	4.6 Discussion
		4.6.1 Comparison with Shard Overlapping in Complete Sharding
		4.6.2 Heterogeneous Blockchain Node
		4.6.3 Multi-Step Transactions
	4.7 Conclusion
	References
5 Sharding-Based Scalable Consortium Blockchain
	5.1 Overview
	5.2 System Model
		5.2.1 Challenges (Goals)
		5.2.2 Our Solution
	5.3 Detailed Cross-Shard Solutions
		5.3.1 Cross-Epoch and Cross-Call
		5.3.2 Partial Cross-Call Merging Strategy
		5.3.3 Replay-Epoch
		5.3.4 Shadow Shard Based Recovery
	5.4 Implementation and Evaluation
		5.4.1 RQ1: Cross-Shard Efficiency
		5.4.2 RQ2: Multi-State Dependency
		5.4.3 RQ3: Consumption of Transaction Atomicity
		5.4.4 RQ4: System Availability
	5.5 Conclusion
	References
6 State Sharding for Permissioned Blockchain
	6.1 Overview
	6.2 System Overview
	6.3 Distributed Blockchain Transaction Structure
		6.3.1 Shard Flag and Stage Flag
		6.3.2 JUMP Operation
		6.3.3 Atomicity
	6.4 Distributed State Update Sharding
		6.4.1 Sharding Strategy
		6.4.2 Round-Based Distributed State Update
	6.5 Evaluation
		6.5.1 Implementation
		6.5.2 Experimental Setup and Workloads
		6.5.3 RQ1: Bottleneck of Current Blockchain
		6.5.4 RQ2: Scalability of Aeolus
		6.5.5 RQ3: Comparison with Other Systems
		6.5.6 RQ4: Impact of Workload
	6.6 Related Work
	6.7 Discussion
	6.8 Conclusion and Future Work
	References
7 Elastic Resource Allocation in Sharding-Based Blockchains
	7.1 Overview
	7.2 Related Work
	7.3 System Model and Problem Formulation
		7.3.1 Sharding-Based Permissioned Blockchain
		7.3.2 Blockchain Shards
		7.3.3 Arrived Transactions in Each Network Shard
		7.3.4 Threat Model of Bursty-TX Injection Attack
		7.3.5 Problem Formulation
	7.4 Dynamic Resource-Allocation Algorithm
		7.4.1 Algorithm Design
		7.4.2 Algorithm Analysis
			7.4.2.1 Upper Bound of System Objective
			7.4.2.2 Upper Bound of the Queue Length of Shards
	7.5 Performance Evaluation
		7.5.1 Basic Settings for Numerical Simulation
		7.5.2 Metrics
		7.5.3 Baselines
		7.5.4 Performance Analysis
			7.5.4.1 Effect of Tuning Parameter V
			7.5.4.2 Performance Comparison with Baselines
			7.5.4.3 Continued Bursty-TX Injection Attacks Against All Shards
			7.5.4.4 Drastic Bursty-TX Injection Attack Against A Single Shard
	7.6 Conclusion
	References
8 Dynamic Sharding: A Trade-OFF Between Securityand Scalability
	8.1 Overview
	8.2 System Overview
		8.2.1 System Model
		8.2.2 Threat Model
		8.2.3 DRL-Based Sharding Model
	8.3 Adaptive Ledger Protocol
		8.3.1 State Block
		8.3.2 Ledgers Merging and Splitting
		8.3.3 Shards Formation
	8.4 Blockchain Sharding System Analysis
		8.4.1 Definitions of Blockchain Sharding System
		8.4.2 Performance Analysis
		8.4.3 Security Analysis
		8.4.4 Problem Formulation
	8.5 DRL-Based Dynamic Sharding Framework
		8.5.1 DRL Model Design
		8.5.2 DRL Training Methodology
		8.5.3 DRL Deployment Methodology
	8.6 Evaluation
		8.6.1 Convergence Analysis
		8.6.2 Security and Latency Performance
		8.6.3 Throughput Comparison with the Baselines
	8.7 Conclusion
	References
9 A Scalable and Secure Framework for 5G Networks Applications
	9.1 Overview
	9.2 Preliminaries of Distributed Machine Learning in 5G Networks
		9.2.1 Consensus Protocols for Decentralized Learning in 5G
		9.2.2 Configurations of Distributed Machine Learning
	9.3 State-of-the-Art Studies of Byzantine-Resilient Machine Learning
		9.3.1 Byzantine-Resilient Machine Learning
		9.3.2 Byzantine Protection on Gradients
		9.3.3 Risks of Decentralization
	9.4 Our Proposal—PIRATE: A Machine Learning Framework Based on Sharding Technique
		9.4.1 Overview of PIRATE
		9.4.2 Permission Control
		9.4.3 Sharding-Based Blockchain Protection Towards Decentralized Distributed-Learning
		9.4.4 Intra-Committee Consensus
		9.4.5 Committee-Wise Ring Allreduce
		9.4.6 Security and Complexity Analysis
			9.4.6.1 Security Analysis of Convergence Attack
			9.4.6.2 Security Analysis of Take-Over Attack
			9.4.6.3 Computation Overhead
		9.4.7 Applications
			9.4.7.1 Decentralized Federated Learning
			9.4.7.2 Big Data Analysis for Consortium Blockchains
	9.5 Case Study
		9.5.1 Security
		9.5.2 Performance Evaluation
	9.6 Open Issues
	9.7 Conclusion and Future Work
	References