Survey of adaptive containerization architectures for HPC

TL;DR

This survey reviews adaptive containerization architectures tailored for HPC, highlighting their role in improving deployment speed, security, and integration with workload managers, amidst growing adoption in supercomputing environments.

Contribution

It provides a comprehensive analysis of container tools, architectures, and integration scenarios specific to HPC, addressing a gap in existing research predominantly focused on cloud environments.

Findings

Containers enhance HPC workflow portability and reproducibility.

Adaptive architectures accelerate application deployment on HPC systems.

Integration with workload managers is crucial for effective HPC containerization.

Abstract

Containers offer an array of advantages that benefit research reproducibility and portability across groups and systems. As container tools mature, container security improves, and High-performance computing (HPC) and cloud system tools converge, supercomputing centers are increasingly integrating containers in their workflows. The technology selection process requires sufficient information on the diverse tools available, yet the majority of research into containers still focuses on cloud environments. We consider an adaptive containerization approach, with a focus on accelerating the deployment of applications and workflows on HPC systems using containers. To this end, we discuss the specific HPC requirements regarding container tools, and analyze the entire containerization stack, including container engines and registries, in-depth. Finally, we consider various orchestrator and HPC workload manager integration scenarios.