Blockchain Systems -- StateSnap: A State-Memorized P2P Storage Network for Blockchain NFT Content Data -- Towards Requester-Provider Bilateral Utility Maximization and Collision Resistance in Blockchain-based Microgrid Energy Trading -- Evaluating the Parallel Execution Schemes of Smart Contract Transactions in Different Blockchains: An Empirical Study -- Misbehavior Detection in VANET Based on Federated Learning and Blockchain -- Data Science -- ABE-AC4DDS: An Access Control Scheme based on Attribute-Based Encryption for Data Distribution Service -- An Interactive Visual System for Data Analytics of Social Media -- Auto-Recon: an Automated Network Reconnaissance System Based on Knowledge Graph -- INGCF: An Improved Recommendation Algorithm Based on NGCF -- Distributed and Network-based Computing -- AutoFlow: Hotspot-Aware, Dynamic Load Balancing for Distributed Stream Processing -- BMTP: Combining Backward Matching with Tree-based Pruning for Large-scale Content-based Pub/Sub Systems -- AF-TCP: Traffic Congestion Prediction at Arbitrary Road Segment and Flexible Future Time -- Collaborative QoS Prediction via Context-Aware Factorization Machine -- TDCT: Target-Driven Concolic Testing Using Extended Units by Calculating Function Relevance -- Multi-Task Allocation Based on Edge Interaction Assistance in Mobile Crowdsensing -- A Variable-way Address Translation Cache for the Exascale Supercomputer -- PPCTS: Performance Prediction-based Co-located Task Scheduling in Clouds -- Edge Computing and Edge Intelligence -- Risk-Aware Optimization of Distribution-based Resilient Task Assignment in Edge Computing -- Budget-aware Scheduling for Hyperparameter Optimization Process in Cloud Environment -- Workload Prediction and VM Clustering Based Server Energy Optimization in Enterprise Cloud Data Center -- Soft Actor-Critic-Based DAG Tasks Offloading in Multi-Access Edge Computing with Inter-User Cooperation -- Service Dependability and Security Algorithms -- SensorData Normalization among Heterogeneous Smartphones for Implicit Authentication -- Privacy-Preserving and Reliable Federated Learning -- Security Authentication of Smart Grid Based on RFF -- SEPoW: Secure and Efficient Proof of Work Sidechains -- A Spectral Clustering Algorithm Based on Differential Privacy Preservation -- A Compact Secret Image Sharing Scheme Based on Flexible Secret Matrix Sharing Scheme -- Robust Multi-Model Personalized Federated Learning via Model Distillation -- V-EPTD: A Verifiable and Efficient Scheme for Privacy-preserving Truth Discovery -- FedSP: Federated Speaker Verification with Personal Privacy Preservation -- Security Performance Analysis for Cellular Mobile Communication System with Randomly-Located Eavesdroppers -- Short and Distort: A New Manipulation as Opposed to Pump and Dump in the Cryptocurrency Market -- Privacy-preserving Swarm learning Based on Homomorphic Encryption -- Software Systems and Efficient Algorithms -- A modeling and verification method of Modbus TCP / IP protocol -- Completely Independent Spanning Trees in the Line Graphs of Torus Networks -- Adjusting OBSS/PD based on fuzzy logic to improve throughput of IEEE 802.11ax network -- Two-Stage Evolutionary Algorithm Using Clustering for Multimodal Multi-Objective Optimization with Imbalance Convergence and Diversity -- SLA: A Cache Algorithm for SSD-SMR Storage System with Minimum RMWs -- Trajectory Similarity Search with Multi-level Semantics -- Nonnegative Matrix Factorization Framework For disease-related CircRNA prediction -- A Fast Authentication and Key Agreement Protocol Based on Time-sensitive Token for Mobile Edge Computing -- Ferproof: A Constant Cost Range Proof Suitable for Floating-Point Numbers -- NEPG: Partitioning Large-Scale Power-Law Graphs -- Accelerating DCNNs via Cooperative Weight/Activation Compression -- PFA: performance and fairness-aware LLC partitioning method -- SGP: A Parallel Computing Framework for Supporting Distributed Structural Graph Clustering -- LIDUSA -- A Learned Index Structure for Dynamical Uneven Spatial Data -- Why is Your Trojan NOT Responding? A Quantitative Analysis of Failures in Backdoor Attacks of Neural Networks -- OptCL: A Middleware to Optimise Performance for High Performance Domain-Specific Languages on Heterogeneous Platforms.

The three volume set LNCS 13155, 13156, and 13157 constitutes the refereed proceedings of the 21st International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2021, which was held online during December 3-5, 2021. The total of 145 full papers included in these proceedings were carefully reviewed and selected from 403 submissions. They cover the many dimensions of parallel algorithms and architectures including fundamental theoretical approaches, practical experimental projects, and commercial components and systems. The papers were organized in topical sections as follows: Part I, LNCS 13155: Deep learning models and applications; software systems and efficient algorithms; edge computing and edge intelligence; service dependability and security algorithms; data science; Part II, LNCS 13156: Software systems and efficient algorithms; parallel and distributed algorithms and applications; data science; edge computing and edge intelligence; blockchain systems; deept learning models and applications; IoT; Part III, LNCS 13157: Blockchain systems; data science; distributed and network-based computing; edge computing and edge intelligence; service dependability and security algorithms; software systems and efficient algorithms.