By João M. Damas
Last month in Amsterdam, the first CompBioMed Containerisation Meeting gathered together some of the world-leading parties in the field of container technologies to discuss the present and the future of these technologies in Cloud and High Performance Computing research and commercial applications. The organizing committee invited me to talk in the meeting, where I shared the recent developments that have been happening in Acellera related with containerisation for reproducible deployment of biomedical applications and workflows in diverse computing infrastructures.
The rate that computational state-of-the-art methods applied to biomedicine take to reach the researcher (i.e. the end user) and impact research is still low. This bottlenecks, for example, the speed of identification of new targets and development of new drugs. In Acellera, we have been focusing on building robust pipelines and efficient workflows from these computational methods and making them easily available to the researcher through the PlayMolecule platform we have been developing. PlayMolecule aims to bring easy-to-use and intuitive applications that can work from anyone’s web browser and run on diverse computing infrastructure. From small computer clusters to HPC, there is no necessity to know about the infrastructural details behind it, allowing the researcher to focus on their research projects.
Containerisation technologies play an essential and central role in this process, as their inherent isolation allows the deployment of any workflow or application in any computing node with little resistance in what regards system management and administration. Furthermore, containerisation allows scaling up the number of applications with different requirements without any impact to the infrastructure. In the meeting, these and other subjects that concern containers have been put to discussion, such as the definition of technology standards, data security and privacy, and the balance between performance and portability.