How to assess the digitization and digital effort: A framework for Digitization Footprint (Part 1)
Qiang Huang, Xu Wang, Qi Gao, Alberto Carraro, Andrea Pezzuolo, Francesco Marinello
Abstract
With the advancement of digitization, data is being collected, transmitted and used in a wide range of domains, and ICT estimates that by 2035, the world will have generated 180 ZB of data, and that in three hundred years’ time, the number of bits generated worldwide may exceed the number of atoms on the planet, and virtual environments (VE) may become unsustainable. In this context, Digitization Footprint (DF) has been proposed to support the analysis of VE sustainability by parameterizing the amount of digital information. At present, the lack of methods for DF evaluation has led to the fact that the current research on DF only stays in the quantification of data volume, and does not quantify the time, effort (costs invested for data storage, processing or transfer). To address this problem, this paper proposes a DF-LCA framework based on Data Processing Units (DPUs), through which the DF of an information system can be scientifically and completely quantified. Considering the complexity of the information system’s internal situation (data transfer, module coupling), it is a challenge to monitor the internal data as well as the metrics. Therefore, by abstracting the information system’s data manipulation process into a DPU, which is a logical unit, it is possible to obtain the information system’s DF by monitoring it from the outside without having to study the information system’s complex interior, which greatly simplifies the difficulty of quantifying the DF. In reality, data may be processed multiple times, and the processing may be distributed over multiple DPUs, and quantifying the DF of a single DPU does not fully satisfy the needs of DF assessment. To address this problem, this paper improves the LCA to DF-LCA by defining the topology between DPUs, clarifying the network relationship of data transmission from DPU to DPU, and finally obtaining the data link of the output data. Finally the DPUs on the data link of this data are monitored with indicators, and the DF of this data can be obtained. In order to solve the problem of transmission error on the link, this paper is based on the existing LCA method, and improves the data bloodline in the big data, which achieves the function of tracking the data transmission problem on the link. In addition, this paper optimizes the hierarchical structure of LCA to make it more applicable to the analysis of DPU topology results of information systems, and finally integrates the three dimensions of performance, energy consumption and value into the index system of DF-LCA. Through the above methods, this paper establishes a generic DF assessment framework and provides basic model support for DF standardization.