Held in conjunction with ISC'21
Held virtually as part of the ISC'21 conference platform on Friday, July 2, 2021, 2 pm to 6 pm Central European Time Zone. To register please follow this link.
2.00pm - 3.30pm Session I Chair: François Tessier (INRIA)
2.00pm - 2.45pm Invited talk: “Met Office supercomputing – cloud is forecast” by Nic Bellingham, Supercomputing Technology Lead (UK Met Office )
Abstract: In April, the Met Office announced a multimillion-pound agreement with Microsoft for the provision of its supercomputing capability for the next 10 years. As the Met Office and Microsoft work together, the collaboration is expected to deliver both scientific and technological innovation. In this talk, Nic will explore what this might mean for the future of Met Office supercomputing.2:45pm - 3:30pm Invited talk: "OpenStack and the Software-Defined Supercomputer" by Stig Telfer (StackHPC)
Abstract: In recent years the proposition of cloud-native supercomputing has matured and is a compelling alternative to conventional HPC infrastructure. Cloud means many things, and on-premise private cloud infrastructure covers the full range of design choices. In this talk, Stig will present recent technical work that shifts the balance in design trade-offs, and may change the way people think about deploying software-defined infrastructure for supercomputing.3:30pm - 4:00pm Break
4:00pm - 6:00pm Session II Chair: David Hancock (Indiana University)
4:00pm - 4:30pm Paper: "A Middleware Supporting Data Movement in Complex and Software-Defined Storage and Memory Architectures" by Christopher Haine (HPE EMEA Research Lab)
Abstract: Among the broad variety of challenges that arise from workloads in a converged HPC and Cloud infrastructure, data movement is of paramount importance, especially oncoming exascale systems featuring multiple tiers of memory and storage. While the focus has, for years, been primarily on optimizing computations, the importance of improving data handling on such architectures is now well understood. As optimization techniques can be applied at different stages (operating system, run-time system, programming environment, and so on), a middleware providing a uniform and consistent data awareness becomes necessary. In this paper, we introduce a novel memory- and data-aware middleware called Maestro, designed for data orchestration.4:30pm - 5:00pm Paper: "Automation for Data-Driven Research with the NERSC Superfacility API" by Bjoern Enders and Gabor Torok (NERSC, LBNL)
Abstract: The Superfacility API brings automation to the use of High Performance Computing (HPC) systems. Our aim is to enable scientists to reliably automate all the tasks they run at the National Energy Research Scientific Computing Center (NERSC), removing human intervention from the process of transferring, analyzing, and managing data. In this paper, we describe the science use cases that drive the API design, our schema of API endpoints, and our implementation strategy, including its authentication and authorization model. We also discuss future plans, working towards our vision of supporting entirely automated experiment-network-HPC workflows.5:00pm - 6:00pm Panel: A 5 Years' Retrospective on Convergence of HPC and Cloud Technologies for Scientific Computing
Panelists: Arjun Shankar (ORNL), Stig Telfer (StackHPC), Andrew Jones (Microsoft Azure), Sadaf Alam (CSCS-ETHZ)
Exascale computing initiatives are expected to enable breakthroughs for multiple scientific disciplines. Increasingly these systems may utilize cloud technologies, enabling complex and distributed workflows that can improve not only scientific productivity, but accessibility of resources to a wide range of communities. Such an integrated and seamlessly orchestrated system for supercomputing and cloud technologies is indispensable for experimental facilities that have been experiencing unprecedented data growth rates. While a subset of high performance computing (HPC) services have been available within a public cloud environments, petascale and beyond data and computing capabilities are largely provisioned within HPC data centres using traditional, bare-metal provisioning services to ensure performance, scaling and cost efficiencies. At the same time, on-demand and interactive provisioning of services that are commonplace in cloud environments, remain elusive for leading supercomputing ecosystems. This workshop aims at bringing together a group of experts and practitioners from academia, national laboratories, and industry to discuss technologies, use cases and best practices in order to set a vision and direction for leveraging high performance, extreme-scale computing and on-demand cloud ecosystems. Topics of interest include tools and technologies enabling scientists for adopting scientific applications to cloud interfaces, interoperability of HPC and cloud resource management and scheduling systems, cloud and HPC storage convergence to allow a high degree of flexibility for users and community platform developers, continuous integration/deployment approaches, reproducibility of scientific workflows in distributed environment, and best practices for enabling X-as-a-Service model at scale while maintaining a range of security constraints.
This workshop will cover topics related to interoperability of supercomputing and cloud computing, networking and storage technologies that are being leveraged by use cases and research infrastructure providers with a goal to improve productivity and reproducibility of extreme-scale scientific workflows:
Virtualization for HPC e.g. virtual machines, containers, etc.
Storage systems for HPC and cloud technologies
On-demand and interactivity with performance, scaling and cost efficiencies
Resource management and scheduling systems for HPC and cloud technologies
Software defined infrastructure for high-end computing, storage and networking
Application environment, integration and deployment technologies
Secure, high-speed networking for integrated HPC and cloud ecosystems
Extreme data and compute workflows and use cases
Research infrastructure deployment use cases
Resiliency and reproducibility of complex and distributed workflows
Isolation and security within shared HPC environments
X-as-a-Service technologies with performance and scalability
Workflow orchestration using public cloud and HPC data centre resources
Authentication, authorization and accounting interoperability for HPC and cloud ecosystems
Workforce development for integrated HPC and cloud environments
Important Dates
Extended abstract submission deadline: April 9, 2021 April 16, 2021
Author notification: April 30, 2021 May 10, 2021
Full paper (conference ready) deadline: June 28, 2021
Workshop date and time: July 2, 2021
Camera-ready papers for Proceedings Deadline: early August 2021
Please refer to ISC 2021 workshop for additional timelines and details: https://www.isc-hpc.com/workshops-2021.html
Paper Submission
Submissions will be done through EasyChair: https://easychair.org/conferences/?conf=isc21supercompcloud
All submissions must be in English and should follow the LNCS style (single column format) using either the LaTeX document class or Word template. Incorrectly formatted papers will be excluded from the reviewing process. Submission must be made as a single PDF file which includes all figures and references. Papers should not be submitted in parallel to any other conference or journal.
The Workshop will accept up to 5 pages extended abstract by the extended abstract submission deadline (see Important Dates above). Authors of selected abstracts, after a peer-reviewed process, will be invited to submit a full paper before the workshop (12 pages maximum in LNCS format, including figures and references) and to present their work on July 2, 2021. A camera-ready deadline will allow to slightly adjust their contributions based on the feedback given during the workshop. Full papers will be published in the ISC 2021 workshop proceedings with Springer. The proceedings will appear as post-conference workshop proceedings. Please refer to ISC Workshop https://www.isc-hpc.com/workshops-2021.html for further details.
Organizing Committee (supercompcloud@googlegroups.com)
Sadaf Alam, Swiss National Supercomputing Centre, Switzerland
François Tessier, INRIA, France
David Y. Hancock, Indiana University, USA
Winnona G. Snapp-Childs, Indiana University, USA
John M. Lowe, Indiana University, USA
Gabriel Antoniu, Inria
Daniel Sokolowski, Technical University of Darmstadt
Julian Pistorius, University of Arizona
Mallikarjun Shankar, ORNL
Steve Quenette, Monash University
Kento Sato, RIKEN R-CCS
Pierre Riteau, StackHPC Ltd
Cerlane Leong, CSCS
Preston Smith, Purdue University
Bruce D'Amora, IBM
Committee members are currently being solicited and confirmed. If you would like to participate in the committee please contact the workshop organizers at supercompcloud@googlegroups.com.