Dataset for the paper "An Empirical Study on Estimating Energy Consumption From Software Traces"

Related Publication Paper Title: An Empirical Study on Estimating Energy Consumption From Software Traces Authors: Phuong Nguyen, Université de Pau et des Pays de l’Adour, Pau, France - [e-mail] Adel Noureddine, Sorbonne Université, CNRS, LIP6 & Université Paris Nanterre, Paris, France - [e-mail] Congduc Pham, Université de Pau et des Pays de l’Adour, Pau, France - [e-mail] Conference: IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT 2025). Dataset Description This dataset contains collected traces, extracted metrics, and statistical correlation analyses used in the experiments described in the above paper. It was created to support reproducibility and enable further analysis of energy estimation from software traces. The data includes measurements from both synthetic and realistic workloads, covering diverse application scenarios and experimental conditions. Main Folder Overview ├── SyntheticWorkloads/ │ ├── 100ms/ │ │ ├── d4r1PerformanceTest_ubuntu/ # CPU-only tests │ │ │ ├── summary_correlation_results.csv │ │ │ ├── run_no_1/, run_no_2/, ... │ │ │ │ ├── hardware_measurevalues_results.csv │ │ │ │ └── hardware_correlation_results.csv │ │ ├── d4r2PerformanceTest_ubuntu/ # Memory-only tests │ │ │ ├── summary_correlation_results.csv │ │ │ └── run_no_1/, run_no_2/, ... │ │ └── d4r3PerformanceTest_ubuntu/ # Combined CPU & memory tests │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ └── 1s/ │ ├── d4r1PerformanceTest_ubuntu/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ ├── d4r2PerformanceTest_ubuntu/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ └── d4r3PerformanceTest_ubuntu/ │ ├── summary_correlation_results.csv │ └── run_no_1/, run_no_2/, ... │ ├── RealisticWorkloads-HotelReservation/ │ ├── summary_correlation_results.csv │ ├── run_no_1/, run_no_2/, ... │ │ ├── dk_measurevalues_results.csv │ │ └── dk_correlation_results.csv │ └── test.jmx │ ├── RealisticWorkloads-Maven/ │ ├── mvn_test/ │ │ ├── johnzon/ │ │ │ ├── summary_correlation_results.csv │ │ │ └── run_no_1/, run_no_2/, ... │ │ ├── JSON-java/ │ │ │ ├── summary_correlation_results.csv │ │ │ └── run_no_1/, run_no_2/, ... │ │ ├── json-simple/ │ │ │ ├── summary_correlation_results.csv │ │ │ └── run_no_1/, run_no_2/, ... │ │ ├── jackson-databind/ │ │ │ ├── summary_correlation_results.csv │ │ │ └── run_no_1/, run_no_2/, ... │ │ └── esson/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ └── mvn_build/ │ ├── Activiti/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ ├── litemall/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ ├── jitwatch/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ ├── languagetool/ │ │ ├── summary_correlation_results.csv │ │ └── run_no_1/, run_no_2/, ... │ └── clojure/ │ ├── summary_correlation_results.csv │ └── run_no_1/, run_no_2/, ... │ ├── RealisticWorkloads-Daytrader/ │ ├── daytrader7/ │ │ ├── summary_correlation_results.csv │ │ └── daytrader7_run_no1/, daytrader7_run_no2/, ... │ │ ├── daytrader7_all.csv │ │ └── correlation_results.csv │ └── daytrader10/ │ ├── summary_correlation_results.csv │ └── daytrader10_run_no1/, daytrader10_run_no2/, ... │ ├── daytrader10_all.csv │ └── correlation_results.csv │ └── RealisticWorkloads-CustomRESTApp/ ├── Exp-i/ │ ├── summary_correlation_results.csv │ └── exp1_run_no1/, exp1_run_no2/, ... │ ├── exp1_all.csv │ └── correlation_results.csv ├── Exp-ii/ │ ├── summary_correlation_results.csv │ └── exp2_run_no1/, exp2_run_no2/, ... │ ├── exp2_all.csv │ └── correlation_results.csv └── Exp-iii/ ├── summary_correlation_results.csv └── exp3_run_no1/, exp3_run_no2/, ... ├── exp3_all.csv └── correlation_results.csv SyntheticWorkloads/ Contains controlled experiments with synthetic workloads. Metrics and energy measurements were collected at: 100 ms intervals 1 s intervals Each subfolder (d4r1PerformanceTest_ubuntu, d4r2PerformanceTest_ubuntu, d4r3PerformanceTest_ubuntu) includes: summary_correlation_results.csv - Average of Pearson correlations across runs. run_no_*/ - Individual test runs containing: hardware_measurevalues_results.csv - Extracted metrics summaries. hardware_correlation_results.csv - Correlation analysis results between energy consumption and system metrics. Logs and trace files for each execution. RealisticWorkloads-HotelReservation/ Contains experiments based on a Hotel Reservation application. summary_correlation_results.csv - Average of Pearson correlations across runs. run_no_*/ - Individual test runs containing: dk_measurevalues_results.csv - Extracted metrics summaries. dk_correlation_results.csv - Correlation analysis results between energy consumption and system metrics. Logs and trace files for each execution. test.jmx - The JMeter configuration used to generate the workload. RealisticWorkloads-Maven/ Contains realistic Maven build and test workloads. mvn_test/ johnzon/ JSON-java/ json-simple/ jackson-databind/ esson/ mvn_build/ Activiti/ litemall/ jitwatch/ languagetool/ clojure/ Each project folder includes: summary_correlation_results.csv - Average of Pearson correlations across runs. run_no_*/ - Individual test runs containing: mvn_measurevalues_results.csv - Extracted metrics summaries. mvn_correlation_results.csv - Correlation analysis results between energy consumption and system metrics. Logs and trace files for each execution. RealisticWorkloads-Daytrader/ Contains results from experiments using the Daytrader benchmark (versions 7 and 10). Each version includes: summary_correlation_results.csv - Average correlation summary. daytrader*_run_no*/ - Run folders containing: <workload>_all.csv - Extracted metrics summaries. correlation_results.csv - Correlation analysis results between energy consumption and system metrics. Logs and trace files for each execution. RealisticWorkloads-CustomRESTApp/ Contains experiments with a custom RESTful web application developed for this study, organized into three experimental series: Exp-i/ Exp-ii/ Exp-iii/ Each experiment folder includes: summary_correlation_results.csv - Average correlation summary across runs. Multiple run folders (e.g., exp1_run_no1/, exp2_run_no3/, exp3_run_no8/) containing: exp*_all.csv - Extracted measurement data. correlation_results.csv - Pearson correlation results between energy and system metrics. Logs and trace files for each execution.