Abstract
The flexibility and portability commonly known from Clouds provide many benefits to users, software developers, administrators and data-center owners. With emerging technologies addressing today’s major bottleneck of virtualization technologies, the virtual I/O, incentives for adoption of virtualization in HPC infrastructures arise. The advantages are manyfold for virtualization in HPC. Users can be served with flexible customized environments. Software developers can package applications with best matching dependencies. Administrators can upgrade or change their HPC infrastructure, i.e. the operating system, without impact on the applications served. Data-center owners can serve conflicting user groups and increase overall resource utilization by consolidating workloads with opposite characteristics. vTorque is an non-intrusive approach to introduce virtualization capabilities to the PBS based batch-system resource manager Torque. For traditional HPC infrastructures it enables cloud-like features, e.g. flexibility and portability, while maintaining the ability to run jobs on bare metal. vTorque further integrates available optimizations for virtual I/O throughout the whole stack, from the hypervisor to the guest level, as optional components.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Adaptive Computing, Inc.: Torque. Documentation. http://docs.adaptivecomputing.com/torque/6-1-2/adminGuide/torque.htm. Cited 01 Jul 2019
Adaptive Computing, Inc.: Torque. qsub Documentation. http://docs.adaptivecomputing.com/torque/6-1-2/adminGuide/torque.htm#topics/torque/commands/qsub.htm
Adaptive Computing, Inc.: Torque. qmgr Documentation. http://docs.adaptivecomputing.com/torque/6-1-2/adminGuide/torque.htm#topics/torque/commands/qmgr.htm
EU Project MIKELANGELO: IOCM. Technology Description. https://www.mikelangelo-project.eu/technology/iocm-io-core-manager/. Cited 01 Jul 2019
EU Project MIKELANGELO: UNCLOT. Technology Description. https://www.mikelangelo-project.eu/technology/UNCLOT-unikernel-cross-level-communication-optimisation/. Cited 01 Jul 2019
EU Project MIKELANGELO: vRDMA. Technology Description. https://www.mikelangelo-project.eu/technology/iocm-io-core-manager/. Cited 01 Jul 2019
EU Project MIKELANGELO: Snap-Telemetry. Technology Description. https://www.mikelangelo-project.eu/technology/full-stack-monitoring/. Cited 01 Jul 2019
EU Project MIKELANGELO: Snap-Telemetry. Technology Description. https://www.mikelangelo-project.eu/technology/universal-unikernel-osv/. Cited 01 Jul 2019
RedHat: Cloud Datasources. NoCloud. https://cloudinit.readthedocs.io/en/latest/topics/datasources/nocloud.html. Cited 01 Jul 2019
LibVirt: Documentation. Domain XML format. https://libvirt.org/formatdomain.html. Cited 01 Jul 2019
Struckmann, N.: Log4bsh. Documentation. https://github.com/mikelangelo-project/log4bsh/blob/master/README.md. Cited 01 Jul 2019
Acknowledgements
vTorque has been developed within project MIKELANGELO—MIcro KErneL virtualizAtioN for hiGh pErformance cLOud and hpc systems, Co-funded by the European Commission under Horizon 2020 Framework Program of the European Union as H2020-ICT-07-2014: Advanced Cloud Infrastructures and Services program.
RIA project no. 645402, from January 2015 to December 2017.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Struckmann, N. (2020). vTorque: Introducing Virtualization Capabilities to Torque. In: Resch, M., Kovalenko, Y., Bez, W., Focht, E., Kobayashi, H. (eds) Sustained Simulation Performance 2018 and 2019. Springer, Cham. https://doi.org/10.1007/978-3-030-39181-2_12
Download citation
DOI: https://doi.org/10.1007/978-3-030-39181-2_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-39180-5
Online ISBN: 978-3-030-39181-2
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)