Testing the DiFX VLBI correlator in a cloud service infrastructure

This study explores the possibility of using cloud infrastructure to deploy the DiFX software correlator for creating a backup system in case of emergencies. The proposed approach involves running DiFX on virtual servers provided by a cloud service, allowing the reduction of financial and time resources. The DiFX software correlator is a primary tool for processing data obtained through Very Long Baseline Interferometry (VLBI). Running such software on cloud provider servers will enable the scaling of the data correlation system. For the experiment, the automation tools Ansible (written in Python) and Terraform were used. Ansible was utilized to deploy the necessary software on a pre-created virtual machine (VM) with Linux, while Terraform managed the provisioning of the virtual machine in the cloud environment. The study analyzes the performance and scalability of DiFX in the cloud. Additionally, the advantages and limitations of deploying DiFX in the cloud in terms of cost, flexibility, and ease of use were evaluated. This paper presents the results of testing and using cloud technologies for running DiFX and justifies the relevance of developing such software.