Software-Driven Strategies to Reduce the Energy Footprint of Data Centers: A Systematic Literature Review
Phase 1: Initial Search & Preprocessing
This folder contains the spreadsheet InitialSearch_SLR_Vos_2021.xlsx. That comprises of:
- Sheet 1 (InitialSearchResults): Articles are ordered according to Google Scholar rank.
- Sheet 2 (RandomizeOrder): Order of articles is randomized by Google Spreadsheet built-in functionality Randomise range.
- Sheet 3 (RemoveCitations): Search results that consist of an empty citation (i.e. the result does not link to an actual article) are removed.
- Sheet 4 (RemovePatents): Google Patents are removed as these are non-scientific articles that are not peer-reviewed.
Phase 2: Selection, Extraction, & Synthesis
This folder contains the spreadsheet Selection&Extraction&Synthesis_SLR_Vosetal_2021.xlsx. That comprises of:
- Sheet 1 (Selection): first, selection based on title and abstract. Then, classification of the primary studies based on the full articles of the selected studies. The selection criteria are explained in detail in the report.
- Sheet 2 (Extraction): from the primary studies, data is extracted according to the categories as defined in the report.
- Sheet 3 (Synthesis): we derived labels from the extracted data and classified the articles accordingly.
- Sheet 4 (Other labels): if the label Other is used in the synthesis phase, the concept that cannot be classified is reported here. This allows the dynamic emerging of new labels as well as insight in the concepts behind the Other categories.
- Sheet 5 (Synthesis Categories): an overview of the possible labels and its definition structured per category.
- Sheet 6 (Selected (Automated Query)): an overview of all the studies that are selected based on their title and abstract.
- Sheet 7 (Primary Studies (Automated Query)): an overview of all the primary studies.
- Sheet 8 (Tag Descriptions): an overview of the definitions of the used tags and abbreviations.
Phase 3: Analysis & Visualization
This folder contains the data, python scripts, and figures of the results from the data synthesis and characteristics of the studies.
- data.csv: the extracted labels for each primary study.
- dataviz.ipynb: python script to generate the figures from data.csv.
- results: output from dataviz.ipynb.
- rejected.csv: the characterestics of the queried studies that are rejected as a primary study.
- selected.csv: the characteristics of the primary studies.
- characteristics.ipynb: python script to visualize the characteristics of the rejected and selected studies.
- characteristics: output from characteristics.ipynb.
Final report containing a detailed description of the research method and results.