G. Borges

TL;DR: The design, development and implementation of this software validation process are detailed, and an analysis is performed on several metrics to evaluate the process impact on the stability of the production infrastructure, by capturing malfunctions and other issues at the initial testing phases.

TL;DR: In order to guarantee a high quality and reliable operation of the infrastructure, all UMD software must undergo a release process that covers the definition of the functional, performance and quality requirements, the verification of those requirements and testing in production environments.

Abstract: This paper presents an implementation of a Ceph file system (CephFS) use case at the ARC Center of Excellence for Particle Physics at the Terascale (CoEPP). CoEPP’s CephFS provides a posix-like file system on top of a Ceph RADOS object store, deployed on commodity hardware and without single points of failure. By delivering a unique file system namespace at different CoEPP centres spread across Australia, local HEP researchers can store, process and share data independently of their geographical locations. CephFS is also used as the back-end file system for a WLCG ATLAS user area at the Australian Tier-2. Dedicated SRM and XROOTD services, deployed on top of CoEPP’s CephFS, integrates it in ATLAS data distributed operations. This setup, while allowing Australian HEP researchers to trigger data movement via ATLAS grid tools, also enables local posix-like read access providing greater control to scientists of their data flows. In this article we will present details on CoEPP’s Ceph/CephFS implementation and report performance I/O metrics collected during the testing/tuning phase of the system.

TL;DR: The int.eu.grid project offers advanced brokering mechanisms and user friendly graphical interfaces supporting application steering enabling full interoperability with existing gLite based infrastructures.