IEEE Transactions on Cloud Computing最新文献

{"title":"P$^{2}$ESchain: Patients-Controlled and Privacy-Preserving EHRs Sharing Based on Blockchain","authors":"Shunrong Jiang;Chengcheng Zhang;Jinpeng Li;Haiqin Wu;Muhammad Sohail;Yong Zhou","doi":"10.1109/TCC.2026.3689576","DOIUrl":"https://doi.org/10.1109/TCC.2026.3689576","url":null,"abstract":"The widespread application of Ehealth systems and medical cloud services has had a profound impact on human society. However, centralized electronic health systems have serious security and privacy risks. In addition, because electronic health records (EHRs) rely on centralized or semi-trusted medical cloud storage platforms, patients lose control over their own health data. Fortunately, the integration of blockchain technology can effectively address these issues. But the transparent on-chain data storage may lead to the leakage of sensitive information in EHRs. To mitigate these vulnerabilities, we intoduce a patient-centric privacy-preserving electronic health record sharing scheme based on blockchain-cloud collaboration (P<inline-formula><tex-math>$^{2}$</tex-math></inline-formula>ESchain). In this scheme, encrypted electronic health records are stored in a semi-trusted medical cloud, and the corresponding hash values are recorded on the blockchain. To realize a secure patient-autonomous sharing mechanism, we construct an anonymous authentication scheme to achieve privacy-protected access right delivery through the blockchain. We also design a secure EHR authorization and delegation algorithm based on Revocable Identity-Based Broadcast Proxy Re-Encryption (RIB-BPRE), ensuring that patients can still receive timely treatment when they are unconscious. The security of the scheme is verified through formal security analysis, and it has been deployed on the local chain (Geth) and the Goerli testnet. Evaluation results show that P<inline-formula><tex-math>$^{2}$</tex-math></inline-formula>ESchain has both security and efficiency, and is suitable for EHRs sharing scenarios in cloud environments.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"2033-2044"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enabling Multicasting Supports for O-RAN Virtual Function Deployment in Multi-Cloud Environment","authors":"Minh Hai Dao;Kim Khoa Nguyen","doi":"10.1109/TCC.2026.3700889","DOIUrl":"https://doi.org/10.1109/TCC.2026.3700889","url":null,"abstract":"The proliferation of multi-user applications in next-generation cellular networks sparks the need to simultaneously deploy multicast trees and provisioning services to provide efficient communication. Such a feature allows taking full advantage of the high flexibility of the Open Radio Access Networks (O-RAN) architecture and its open standards, virtualization, and multi-cloud deployment. Enabling multicasting supports in O-RAN, such as 3GPP’s 5G Multicast Broadcast Service capabilities, will unlock and improve a multitude of new O-RAN use cases. Unfortunately, such integration has not yet been clearly defined in O-RAN specifications or fully investigated in prior studies. In this paper, we jointly formulate the Multicast Service Function Chain Embedding (MSE) and O-RAN functional split optimization problems in a multi-cloud environment as an Integer Linear Programming (ILP) model. We address both single and multiple flow request scenarios and propose a Pseudo Knapsack-based heuristic algorithm with polynomial-time complexity, combining feasible search optimization and dynamic programming techniques. Experimental results show that our approach outperforms state-of-the-art baselines, approximating the optimal solution while significantly reducing computational time.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"1785-1801"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Provable Versioned Inter-Cloud Auditing via Block-Level Redactable Blockchain","authors":"Jingting Xue;Ximin Jing;Fagen Li;Peiling Wu;Wenzheng Zhang;Xiaojun Zhang;Yu Zhou","doi":"10.1109/TCC.2026.3718964","DOIUrl":"https://doi.org/10.1109/TCC.2026.3718964","url":null,"abstract":"As a core infrastructure for data governance, trustworthy cloud storage demands secure and efficient auditing mechanisms, which are essential for unlocking data value.However, existing public auditing schemes suffer from critical limitations, including collusion risks introduced by third-party auditors, storage bloat due to the immutability of blockchain, and inefficiencies in version traceability. In this paper, we propose PVICA, a provable versioned inter-cloud auditing scheme to address these challenges. PVICA adopts a decentralized mutual auditing architecture across multiple clouds based on verifiable delay functions, which inherently prevent precomputation attacks and eliminate potential collusion between clouds or between a cloud and its auditor. To mitigate blockchain-induced storage overhead while ensuring controlled redactability, we design a block-level redactable blockchain with double-trapdoor keys control. By separating ephemeral and long-term trapdoor key authorities, PVICA reduces redundant outsourced logs and enables efficient audit updates. Moreover, we integrate RSA accumulators to achieve constant-time cross-version verification, thereby reconciling the conflict between audit log privacy and regulatory compliance. Formal security analysis demonstrates that PVICA achieves key guarantees such as storage correctness and tamper detection. Experimental results show that its block-level redacting mechanism significantly improves audit throughput over traditional approaches, highlighting its practicality for real-world deployment.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"2086-2099"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Edge Microservice Deployment Based on Genetic Algorithm and Hierarchical Fault Tolerance","authors":"Min Yuan;Jiaxin Mu;Hongbing Wang","doi":"10.1109/TCC.2026.3695764","DOIUrl":"https://doi.org/10.1109/TCC.2026.3695764","url":null,"abstract":"Efficient and stable microservice deployment in edge environments remains challenging due to stringent resource constraints and high node failure rates. Existing works typically address isolated optimization factors, while this singular focus results in suboptimal deployment performance. To overcome these limitations, we propose a new deployment approach combining genetic algorithm (GA) with a hierarchical fault-tolerance mechanism. Our approach first formulates microservice deployment problem as a hierarchical optimization problem, sequentially optimizing response time, cost, deployment success rate under constraints of budget and load. Next, we design a genetic algorithm with an elite chromosome generation strategy to reduce search complexity, incorporating a fault-aware mechanism to avoid unreliable nodes. Furthermore, we introduce a three-layer fault-tolerance mechanism based on cloud-edge-end architecture that performs fault prevention at the cloud layer, real-time remediation at the edge layer, and quality evaluation at the end layer. Extensive experiments on real-world data demonstrate that our approach outperforms baseline methods, improving Quality of Service (QoS) by 13.2% and increasing the deployment success rate (DSR) by 22.9%.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"1732-1744"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dual-Timescale Adaptive Model Partitioning, Jamming Power Control, and Resource Allocation Against Multi-Eavesdroppers","authors":"Hao Yuan;Wenhao Fan;Xiongfei Chun;Ruimin Zhang;Kong Ma;Bihua Tang;Yuan’an Liu","doi":"10.1109/TCC.2026.3697788","DOIUrl":"https://doi.org/10.1109/TCC.2026.3697788","url":null,"abstract":"Model partitioning is a key technique for cooperative edge-device inference, balancing computing loads, reducing communication latency, and preserving privacy by transmitting intermediate parameters instead of raw data. However, eavesdroppers may intercept these intermediate parameters during wireless transmission, potentially recovering sensitive raw data. Current model partitioning studies primarily focus on latency and energy optimization, while neglecting security protection in multi-eavesdropper scenarios. To address this, we propose a dual-timescale adaptive model partitioning, jamming power control and resource allocation scheme against multi-eavesdroppers. Specifically, we jointly optimize model partition point selection for each device, jamming power control from the base station to each eavesdropper, computing resource allocation between end devices and the edge server for each tasks, and transmission power control for each device to minimize the overall latency-energy cost. To tackle the high complexity of the optimization problem and the heterogeneity in update frequencies among optimization variables, we propose a Twin Delayed Deep Deterministic Policy Gradient (TD3)-based dual-timescale reinforcement learning algorithm that jointly optimize the long-term model partitioning decision and other four short-term variables. By integrating numerical optimization techniques, the algorithm effectively reduces action space dimensionality and accelerates convergence. Extensive multi-scenario experiments demonstrate that our algorithm achieves the lowest total system cost and outperforms 4 existing advanced schemes.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"1688-1703"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TCformer: Fine-Grained Cloud Workload Prediction via Multi-Scale Convolution and Channel-Aware Temporal Attention","authors":"Yao Lu;Peng Yu;Renyu Zhang;Zekun Sun;Ziheng Suo;Jiayan Gu;Bo Yuan;Lu Liu;Hong Zhong;Jie Cui;Geyong Min","doi":"10.1109/TCC.2026.3700214","DOIUrl":"https://doi.org/10.1109/TCC.2026.3700214","url":null,"abstract":"Accurate prediction of cloud workloads is vital for optimizing resource allocation and reducing energy consumption in data centers. However, cloud workloads exhibit non-stationarity, burstiness, and weak periodicity, which make forecasting highly challenging. Existing time-series models often fail to capture fine-grained short-term fluctuations, tend to oversmooth long-term trends, and struggle to capture dynamic inter-channel dependencies. These limitations hinder their ability to track abrupt spikes and cross-domain signals that are critical in real-world scenarios. To overcome these issues, we propose TCformer, a Transformer-based prediction model that integrates two key components: a multi-scale convolutional embedding module to capture global and local feature variations across temporal cycles adaptively, and a temporal channel attention mechanism to dynamically adjust channel importance for modelling cross-channel dependencies over time. Extensive experiments on three real-world cloud workload datasets covering 27 workload types demonstrate that TCformer consistently outperforms state-of-the-art baselines, achieving superior accuracy in both short-term and long-term forecasting.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"2168-2180"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Inner-Product Functional Encryption With Fine-Grained Revocation for Flexible EHR Sharing","authors":"Yue Han;Jinguang Han;Liqun Chen;Chao Sun","doi":"10.1109/TCC.2026.3688378","DOIUrl":"https://doi.org/10.1109/TCC.2026.3688378","url":null,"abstract":"E-health record (EHR) contains a vast amount of continuously growing medical data and enables medical institutions to access patient health data conveniently. This provides opportunities for medical data analysis, which has important applications in calculating the average age of patients and linear regression inference for epidemics, etc. Since EHR contains sensitive patient information, how to protect patient privacy and enable analysis on EHR data is important and challenging. Traditional public key encryption (PKE) can protect patient privacy, but cannot support flexible selective computation on encrypted EHR data. Functional encryption (FE) allows authorised users to compute function values of encrypted data without releasing other information, hence supporting selective computation on encrypted data. Nevertheless, existing FE schemes do not support fine-grained revocation and update, so they are unsuitable for EHR system. In this paper, to support fine-grained revocation and update, we first propose an inner-product functional encryption with fine-grained revocation (IPFE-FR) scheme, and then apply it to a flexible EHR sharing system. Our scheme possesses the following features: (1) a group manager can revoke a specific function computation of medical institutions on encrypted EHR data, instead of all function computation rights. (2) a revoked medical institution is not allowed to compute the function value of encrypted EHR data not only generated after the revocation, but also generated before the revocation. (3) secret keys issued to the same medical institution are bound together to prevent collusion attacks. The formal definition and security model of the IPFE-FR scheme are proposed. Furthermore, we present a concrete construction and reduce its security to the Learning with Errors (LWE) assumption which is quantum-resistant. Finally, the theoretical analysis and experimental implementation of our scheme are conducted to show its efficiency. Our results show that the proposed scheme can flexibly and securely share EHR data.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"1904-1917"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148903100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"IST: A Stealthy Backdoor Attack for Self-Supervised Learning Using Invisible Structure Triggers","authors":"Shaobo Zhang;Zhoufan Yang;Yimeng Pan;Wenjuan Tang;Kuan Zhang","doi":"10.1109/TCC.2026.3718628","DOIUrl":"https://doi.org/10.1109/TCC.2026.3718628","url":null,"abstract":"Self-supervised learning (SSL) has shown significant advantages by utilizing unlabeled data for training. However, it remains vulnerable to backdoor attacks. Existing attacks on SSL can achieve a high Attack Success Rate (ASR) with visible triggers, but they suffer from a high Attack Success Rate-Baseline (ASR-B) and lack stealthiness. Conversely, using invisible triggers reduces ASR-B but lowers ASR, failing to effectively balance effectiveness and stealthiness. To address this, this paper proposes IST, a stealthy backdoor attack for SSL using invisible structure triggers, which includes two stages: poisoned data generation and backdoor embedding. In the poisoned data generation stage, the attacker extracts and colors the edges of clean images, embedding them as invisible perturbations within the image structure. In the backdoor embedding stage, the attacker enhances the cosine similarity between poisoned data and target-class clean data by minimizing the loss function, indirectly associating the trigger with the target class without requiring data labels. Experiments show that IST demonstrates superior stealthiness, achieving an average ASR-B reduction of 14.85% and 9.06% compared to state-of-the-art methods like BadEncoder and INACTIVE. Furthermore, IST achieves a higher ASR in various downstream tasks and performs well on the Contrastive Language-Image Pre-training (CLIP) model.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"1933-1947"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148903138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TFRL-MD: Transformer-Enhanced Federated Reinforcement Learning for Intelligent Microservice Deployment and Resource Allocation","authors":"Keli Liu;Jing Yang;Xiaoli Ruan;Yuling Chen;Xu Wang;Jianhong Cheng;Yulong Yang;Minyi Guo","doi":"10.1109/TCC.2026.3703396","DOIUrl":"https://doi.org/10.1109/TCC.2026.3703396","url":null,"abstract":"Edge computing provides low-latency execution environments for delay-sensitive microservices by deploying computational resources at the network periphery. However, microservice deployment and resource allocation in geographically distributed, heterogeneous edge environments must simultaneously navigate challenges such as data privacy preservation, heterogeneous resource constraints, and dynamic workload fluctuations. Therefore, we propose TFRL-MD, a Transformer-enhanced federated reinforcement learning framework for privacy-preserving and intelligent microservice deployment and resource allocation. The framework models each edge node as an independent Actor-Critic agent to learn deployment policies via local training, enabling cross-node collaborative optimization without sharing raw environmental data. To improve the quality of knowledge fusion among heterogeneous edge nodes, we design a Transformer-based Embedding-Space Aggregation Mechanism (EASM) that leverages multi-head self-attention to adaptively fuse their intermediate-layer knowledge representations. Furthermore, a dual-channel hybrid objective training mechanism is constructed to synchronously inject the aggregated global representations into the optimization targets of both Actor and Critic networks as differentiable semantic embeddings, thereby smoothly guiding local gradient descent and eliminating the policy oscillations inherent in traditional parameter overwriting.Dynamic simulation experiments, conducted using the EUA dataset from the Australian Central Business District (CBD) region, demonstrate that TFRL-MD outperforms various baselines in both average response time and load balancing, achieving maximum improvements of 46.01% and 61.11%, respectively.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"2195-2208"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mixer: Accelerating Deep Neural Network Training With Hybrid Optimizers","authors":"Ruidong Yan;Jidong Zhai;Rengang Li;Zhiyong Qiu;Liang Jin;Yuyang Jin;Yaqian Zhao;Zhenhua Guo","doi":"10.1109/TCC.2026.3706801","DOIUrl":"https://doi.org/10.1109/TCC.2026.3706801","url":null,"abstract":"Optimization algorithms, or optimizers, are indispensable for accelerating training in deep learning. However, they often suffer from high computational complexity and inefficiency. In this work, we propose a novel approach called Mixer through combining both first-order and second-order optimization approaches to address these challenges. Mixer has three contributions: (1) introducing adaptive algorithm switching conditions, (2) presenting a gradient vector correction strategy to mitigate accuracy degradation, and (3) incorporating a lightweight second-order method using efficient matrix decomposition to reduce computation and memory overhead. We evaluate Mixer’s performance by training networks such as ResNet and BERT with real datasets, comparing them against several state-of-the-art methods. Results demonstrate Mixer outperforms others under the same convergence accuracy, saving 48.86% and 14.06% end-to-end training time compared with SGD and KAISA for ResNet-152, achieving up to 1.95× and 1.16× speedups, respectively. Besides, Mixer saves 30.93% and 22.63% training time, achieving 1.45× and 1.29× improvement over LAMB and KAISA for BERT-large. Simultaneously, Mixer saves at least 17.42% memory footprint over KAISA.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"14 3","pages":"1975-1988"},"PeriodicalIF":5.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}