Paramart : Parallel Resource Allocation Based on Blockchain Sharding for Edge-Cloud Services

Xiaoxu Ren,Minrui Xu,Dusit Niyato,Jiawen Kang,Chao Qiu,Xiaofei Wang

DOI: https://doi.org/10.1109/tsc.2024.3359608

IF: 11.019

2024-01-01

IEEE Transactions on Services Computing

Abstract:Edge computing has evolved to enable mobile applications to run in an efficient and cost-effective manner at explosive-growing edge nodes. Under this paradigm, a new business resource trading market has emerged to provide edge-cloud services, offering a convenient way for mobile users to obtain resources from distributed computing power providers (CPPs). Blockchain, as a promising technology, provides a reliable platform for multi-party resource transactions (TXs), enabling secure and reliable computing services. Notably, the distributed CPPs not only offer mobile services but also act as blockchain nodes to maintain the stability of TXs. In this case, there exist certain bottlenecks in the blockchain-enabled edge-cloud resource market, such as limited scalability, inefficient resource allocation, and large system cost. In this paper, assisted by the permissioned blockchain, we study the fundamental problem of resource allocation by minimizing the system cost to handle mobile services and blockchain TXs in parallel. We first partition the Practical Byzantine Fault Tolerant (PBFT) consensus by hierarchical sharding to improve the scalability and ensure the security of the blockchain system. Next, based on the optimal sharding strategies, we formulate the parallel resource allocation as a multi-scale Lyapunov optimization problem, and develop a dual-alternation actor-critic with an attention mechanism (DA3C) algorithm to solve it. We evaluate the performance of the Paramart using trace-driven experiments. Simulation results demonstrate the superiority of our proposed framework as compared with the benchmark algorithms.

computer science, information systems, software engineering

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

This paper primarily addresses the issue of resource allocation in the resource trading market within the context of the integration of edge computing and blockchain technology. Specifically:

1. **Resource Allocation and Security**: In an edge computing environment, Computing Power Providers (CPPs) not only provide mobile services but also participate as blockchain nodes to maintain transaction security. However, existing blockchain technology faces bottlenecks in scalability and resource allocation efficiency, leading to high system costs.

2. **Application of Blockchain Sharding Technology**: To improve the scalability of the blockchain system and ensure its security, the paper proposes a resource allocation framework called Paramart based on blockchain sharding technology. By optimizing the Practical Byzantine Fault Tolerant (PBFT) consensus mechanism through a hierarchical sharding strategy, higher scalability and security are achieved.

3. **Multi-Scale Resource Allocation**: The paper models the parallel resource allocation problem as a multi-scale Lyapunov optimization problem and proposes a Dual Alternating Actor-Critic algorithm (DA3C) with an attention mechanism to achieve reasonable allocation of resources between mobile services and blockchain transactions while ensuring Quality of Service (QoS).

In summary, this paper aims to address the issues of scalability, security, and cost in the resource trading market within an edge cloud computing environment by optimizing blockchain sharding technology and resource allocation strategies, thereby improving the overall system performance and service quality.