Deploying a Globus Endpoint in a NZ Institution
2019-05-15T00:25:12Z (GMT) by
Using Globus for high performance data transfer service is a frequently discussed topic in the eResearch NZ conference. including workshops and presentations delivered by Globus staff members. However, its uptake in the NZ research community is still not as wide as one might expect. There seems to be a perception that deploying a Globus endpoint is complicated. We would like to change this perception by showcasing the technical details two recently deployments in the NZ research community.
Dan recently deployed a Globus endpoint in AgResearch. It is a single node deployment and the endpoint is attached to the filesystems of existing Linux cluster. He going to cover its technical design that fits the endpoint in a common corporate IT infrastructure environment.
Jose also recently redeployed a Globus endpoint for NeNI’s new platform. It is a clustered deployment to provide the performance required by a leading national facility. He is going to cover how individual functional components can be separated and be deployed in a cluster environment.
Dan and Jose will also present an Ansible Playbook that may be helpful to others in the community who are also considering deploying a Globus endpoint for their institute.
ABOUT THE AUTHOR(S)
Dan has been an active eResearch community member in NZ for a few years. He is currently an HPC consultant working for AgResearh and maintains their smallish Linux cluster and storage. He is passionate about helping researchers to do science by using advanced technologies. When he is not firefighting at work, he enjoys having fancy burgers and donuts with his collaborators and friends.
José Higino moved from Portugal to New Zealand in 2014 to work with NIWA's Supercomputer, Fitzroy, as a High Performance Computing (HPC) Systems Engineer in Wellington. His has a background in electrotechnics and computers engineering and worked for 7 years at IBM Portugal as a software and services IT Specialist. Passionate about troubleshooting, logic, distributed and parallel systems, he followed the roots of HPC and distributed storage. Currently, he is part of the NeSI's Platforms Team supporting the new Cray Supercomputers, the XC50 and CS500/400.