TY - GEN
T1 - Towards an HPC Service Oriented Hybrid Cloud Architecture Designed for Interactive Workflows
AU - Kortas, Samuel
AU - Shaikh, Mohsin Ahmed
N1 - KAUST Repository Item: Exported on 2021-01-06
Acknowledgements: For computer time, this research used the resources of the Supercomputing Laboratory at King Abdullah University of Science & Technology (KAUST) in Thuwal, Saudi Arabia.
PY - 2020
Y1 - 2020
N2 - We detail Ludion, a service-oriented hybrid architecture, well-adapted to launch, monitor and steer interactive services running either on on-premise HPC resource, on user laptop and workstation or in the cloud. Based on AWS server-less components at virtually no cost, Ludion requires no special privileges. It consists of a catalog of services, a dashboard hosted in the cloud and a set of commands to install on the target resources. In the lifetime of a job, a user can register and publish new service on the dashboard. From this unique location, it is possible to trigger basic commands that are forwarded to the corresponding job. In this article, we expose several typical use cases Ludion was designed for and detail its implementation as well as the services already in place relying on this architecture.
AB - We detail Ludion, a service-oriented hybrid architecture, well-adapted to launch, monitor and steer interactive services running either on on-premise HPC resource, on user laptop and workstation or in the cloud. Based on AWS server-less components at virtually no cost, Ludion requires no special privileges. It consists of a catalog of services, a dashboard hosted in the cloud and a set of commands to install on the target resources. In the lifetime of a job, a user can register and publish new service on the dashboard. From this unique location, it is possible to trigger basic commands that are forwarded to the corresponding job. In this article, we expose several typical use cases Ludion was designed for and detail its implementation as well as the services already in place relying on this architecture.
UR - http://hdl.handle.net/10754/666824
UR - https://ieeexplore.ieee.org/document/9307973/
U2 - 10.1109/UrgentHPC51945.2020.00010
DO - 10.1109/UrgentHPC51945.2020.00010
M3 - Conference contribution
SN - 978-0-7381-1063-9
BT - 2020 IEEE/ACM HPC for Urgent Decision Making (UrgentHPC)
PB - IEEE
ER -