| dc.description.abstract | Accurate and fair evaluation of blockchain transactions is essential for ensuring system efficiency, economic fairness, and resilience to manipulation. Traditional evaluation mechanisms, such as Ethereum`s gas-based model, use static opcode pricing and market-driven fees, which do not reflect a transaction`s actual computational footprint. This leads to inefficiencies, unfair prioritisation, and limited scalability in heterogeneous or high-load environments.
This paper introduces the Resource-Aware Transaction Evaluation Model (RATEM), a multidimensional framework designed to quantify transaction cost based on observable system-level metrics, including CPU time, memory operations, I/O volume, and execution latency. The RATEM approach addresses the gap between abstract fee mechanisms and real resource consumption by offering three adaptable submodels: ETOM (Execution-Time-
Only Model), IREM (Instrumented Resource Evaluation Model), and HGEM (Hybrid Gas Evaluation Model). Each submodel supports a different level of metric availability and validator capability, enabling deployment in both constrained and advanced settings.
The paper presents a formal mathematical formulation of each model, discusses their comparative properties, and provides a taxonomy of resource metrics used in transaction evaluation. Experimental insights and theoretical analysis demonstrate that RATEM improves alignment between transaction cost and system load, supports fairer scheduling decisions, and enhances platform-agnostic comparability.
The proposed framework offers a practical foundation for improving blockchain governance and transaction fee mechanisms in Layer-1 and Layer-2 environments. Future directions include machine learning–based weight calibration, testnet deployment, and integration with smart contract validators. RATEM provides a viable pathway toward transparent, efficient, and resource-aware transaction evaluation in modern blockchain ecosystems. | uk |
