Blockchain-based trust management in cloud computing systems: a taxonomy, review and future directions

Wenjuan Li, Jiyi Wu, Jian Cao, Nan Chen, Qifei Zhang, Rajkumar Buyya

Cloud computing has evolved from a specialized service into a critical global infrastructure. However, as organizations migrate sensitive data and core processes to the cloud, they often encounter a "trust crisis." This paper, published in the Journal of Cloud Computing, explores how blockchain technology can mitigate the risks inherent in centralized cloud models by providing transparent, immutable, and decentralized trust management.

The Problem: The Cloud as a "Black Box"

Traditional cloud trust models rely heavily on centralized third-party authorities. While efficient in some respects, these models create several core risks for users:

  1. Loss of Control: Once data is submitted to remote servers, users lose visibility into how it is handled.
  2. Lack of Transparency: Internal operational mechanisms are often hidden, raising concerns about unauthorized privacy manipulation.
  3. Vague Service Level Agreements (SLAs): SLAs are frequently abstract, making it difficult for users to verify compliance or enforce penalties when breaches occur.

Centralized trust centers also represent single points of failure and can become bottlenecks during high-traffic periods. The authors argue that blockchain’s distributed ledger technology offers a viable alternative by replacing opaque centralized authority with decentralized verification.

A Taxonomy of Blockchain-Based Trust

The authors conducted a systematic review of 35 representative studies published between 2017 and 2020. They categorized these approaches into three distinct phases, as illustrated in the taxonomy below.

Phases of blockchain-based trust approaches in cloud computing systems

1. Basic Trust Frameworks

This phase focuses on the foundational elements of security: identity authentication and behavior evaluation. Blockchain acts as a decentralized registry where user identities and historical behavior are recorded immutably.

Blockchain acting as an identity authentication platform across different trust domains

As shown in the mechanism above, blockchain can facilitate inter-domain service interactions by allowing a "Home Cloud Service Provider (CSP)" to store access tokens that can be verified by a "Foreign CSP" without requiring a central broker.

2. Interaction Frameworks and Mechanisms

This phase addresses the dynamic aspects of cloud services, such as SLA enforcement and resource allocation. A key innovation here is the "witness" mechanism, where nodes in the blockchain supervise transactions. Smart contracts automatically trigger rewards or penalties based on real-time performance data, ensuring that SLAs are enforced programmatically rather than just legally.

3. Data Process and Storage

The final phase focuses on data provenance and access control. By recording every data access event on a blockchain, providers can offer users a verifiable audit trail, resolving the "black box" issue of data manipulation.

Novel Architectural Proposals

Beyond the review, the authors propose two frameworks to solve common blockchain limitations like latency and storage "swell":

  • Cloud-Edge Hybrid Framework: To reduce latency, this model moves time-sensitive trust evaluations to Multi-access Edge Computing (MEC) servers. This three-layer approach (Blockchain, Edge, and Cloud) ensures that local interactions are processed quickly while maintaining global trust.
  • Double-Blockchain Structure: This structure splits the workload into two parallel chains. The Trust Authentication Blockchain (TAB) manages static identity and long-term reputation, while the Trading Behavior Blockchain (TBB) records real-time transaction results. This separation prevents a single chain from becoming too large and slow.

Key Results and Insights

The survey reveals a significant shift toward decentralization, with approximately 63% of the reviewed articles successfully implementing fully decentralized models. Performance metrics are also promising; blockchain-based detection algorithms were found to improve trust evaluation accuracy by 5% to 15% compared to traditional methods. In specific high-capacity scenarios, some implementations reduced processing delays to just 20% of the time required by centralized systems.

Future Directions and Limitations

Despite the progress, several challenges remain:

  • The Privacy Paradox: There is a fundamental tension between the transparency of a public ledger and the privacy required for sensitive data. Future research must integrate zero-knowledge proofs to allow for verification without data exposure.
  • Scalability: As transaction histories grow, the storage demands on nodes can become unsustainable. The authors suggest moving toward "Agent-based Intelligent Trust," where AI agents autonomously manage relationships and optimize data storage.
  • Real-world Deployment: Currently, 62.86% of studies rely on simulations. Moving these frameworks from prototypes to large-scale, cross-provider production environments remains the next major hurdle for the industry.

Summary updated

Abstract

Through virtualization and resource integration, cloud computing has expanded its service area and offers a better user experience than the traditional platforms, along with its business operation model bringing huge economic and social benefits. However, a large amount of evidence shows that cloud computing is facing with serious security and trust crisis, and building a trust-enabled transaction environment has become its key factor. The traditional cloud trust model usually adopts a centralized architecture, which causes large management overhead, network congestion and even single point of failure. Furthermore, due to a lack of transparency and traceability, trust evaluation results cannot be fully recognized by all participants. Blockchain is a new and promising decentralized framework and distributed computing paradigm. Its unique features in operating rules and traceability of records ensure the integrity, undeniability and security of the transaction data. Therefore, blockchain is very suitable for constructing a distributed and decentralized trust architecture. This paper carries out a comprehensive survey on blockchain-based trust approaches in cloud computing systems. Based on a novel cloud-edge trust management framework and a double-blockchain structure based cloud transaction model, it identifies the open challenges and gives directions for future research in this field.
BibTeX
@article{li2021blockchain,
  author = {Wenjuan Li and Jiyi Wu and Jian Cao and Nan Chen and Qifei Zhang and Rajkumar Buyya},
  title = {{Blockchain-based trust management in cloud computing systems: a taxonomy, review and future directions}},
  journal = {J. Cloud Comput.},
  year = {2021},
  doi = {10.1186/s13677-021-00247-5},
}
Figure