Univariate vs multivariate prediction for containerised applications auto-scaling:: a comparative study
We have anonymised the repository to follow blind peer review process requirements. You can clone this repository using GUI software available for Windows or macOS at this link.
However, if you are a Linux user or prefer the command line, you can clone this repository using these commands:
$ git clone https://github.com/fedebotu/clone-anonymous-github.git
$ cd clone-anonymous-github
$ python3 src/download.py --url https://anonymous.4open.science/r/univariate-multivariate-prediction-4054/
This paper presents a comparative study assessing univariate and multivariate proactive auto-scaling of containerised applications. A custom-made multivariate auto-scaling tool called Multivariate Forecasting Tool (MFT) was developed and compared with a production-grade univariate system called Predict Kube (PK). Both applications were evaluated using four popular open-source benchmark applications (Daytrader, Online Boutique, Quarkus-HTTP-Demo and Travels).
$ virtualenv venv
$ source venv/bin/activate
$ pip3 install -r requirements.txt
Summary of the main repository files.
| Files | Content description |
|---|---|
| database | It contains the datasets used for training the multivariate and univariate models. |
| experiment_files | It contains the application files, configuration files, monitoring and workload files. |
| knowledge | It contains the trained multivariate models used in the MFT. |
| mft | It contains the MFT source code. |
| model_training | It contains the source code for multivariate model training. |
| results | It contains the results of experiments. |
| process_results | It contains the files for processing the experiment results. |