Multi-Dimensional Resource Evaluation in Blockchain

The traditional gas-based metering model used in Ethereum and similar blockchain platforms simplifies execution cost estimation but fails to capture the comprehensive complexity of smart contract workloads. This paper introduces a novel multidimensional resource metering framework, designed to separately quantify computational load (based on floating-point operations- FLOPs), memory access, and execution latency. The study effectively demonstrates the framework's capability across various tasks, including compute-intensive processes, in-memory data handling, and high-latency tasks through detailed conceptual modelling of matrix multiplication operations. The proposed model advocates for fairer, hardware-based metering strategies, fostering more precise resource optimisation, and aligns well with contemporary trends such as multidimensional gas pricing in Layer-2 scaling solutions. Additionally, this framework offers critical insights into constructing theoretical and practical models for blockchain resource metering, significantly enhancing accuracy in cost attribution. Experimental validation using representative computational workloads, such as matrix multiplications, highlights distinct performance metrics- FLOPs, memory bandwidth, and latency- and contrasts these findings with traditional single-dimensional gas metrics. Ultimately, this research provides a robust foundation for a more equitable and efficient computational resource evaluation in blockchain systems, paving the way for better system scalability, optimisation, and sustainable economic models for future blockchain applications