The complexity of managing cloud applications’ life-cycle increases with the widening of the cloud landscape, as new IT players gain market share. Cloud orchestration frameworks promise to handle such complexity offering user-friendly management tools that help customers to transparently deal with portability and interoperability issues, by hiding away the heterogeneity of the cloud providers’ proprietary interfaces. Regarding the provisioning of cloud-enabled applications, the containerisation paradigm, along with the related micro-services technology, has managed to deliver the portability promise. While most of cloud orchestration frameworks support container-based cluster technologies, a standard-based approach to describe containerised applications still lacks. In this work, we propose TORCH, a TOSCA-based cloud orchestrator capable of interfacing to theoretically any container run-time software by leveraging a standard-aligned and easy-to-use language to describe application requirements. Validation tests run on a small-scale test-bed prove the viability of the proposed solution.
Cluster-Agnostic Orchestration of Containerised Applications
Calcaterra D.;Di Modica G.;Tomarchio O.
2021-01-01
Abstract
The complexity of managing cloud applications’ life-cycle increases with the widening of the cloud landscape, as new IT players gain market share. Cloud orchestration frameworks promise to handle such complexity offering user-friendly management tools that help customers to transparently deal with portability and interoperability issues, by hiding away the heterogeneity of the cloud providers’ proprietary interfaces. Regarding the provisioning of cloud-enabled applications, the containerisation paradigm, along with the related micro-services technology, has managed to deliver the portability promise. While most of cloud orchestration frameworks support container-based cluster technologies, a standard-based approach to describe containerised applications still lacks. In this work, we propose TORCH, a TOSCA-based cloud orchestrator capable of interfacing to theoretically any container run-time software by leveraging a standard-aligned and easy-to-use language to describe application requirements. Validation tests run on a small-scale test-bed prove the viability of the proposed solution.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.