2019
DOI: 10.1007/978-3-030-34356-9_40
|View full text |Cite
|
Sign up to set email alerts
|

The Role of Interactive Super-Computing in Using HPC for Urgent Decision Making

Abstract: Technological advances are creating exciting new opportunities that have the potential to move HPC well beyond traditional computational workloads. In this paper we focus on the potential for HPC to be instrumental in responding to disasters such as wildfires, hurricanes, extreme flooding, earthquakes, tsunamis, winter weather conditions, and accidents. Driven by the VESTEC EU funded H2020 project, our research looks to prove HPC as a tool not only capable of simulating disasters once they have happened, but a… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
4
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
5
2

Relationship

3
4

Authors

Journals

citations
Cited by 7 publications
(5 citation statements)
references
References 9 publications
0
4
0
Order By: Relevance
“…One alternative to using HPC machines is to use the cloud, however, this is not as attractive an option as it may initially seem. In a previous study for VESTEC [39], it was found that the performance of cloud systems tends to be significantly lower than a corresponding HPC machine, mainly due to filesystem performance, the cost was higher, and the cloud is not as elastic as one would hope, with reasonably long startup wait times for large, infrequent compute requirements. Whilst these make it impractical to rely upon a cloud-based resource, it would still be plausible to supplement the HPC machines we federate over with some cloud resources for some jobs/applications.…”
Section: Discussionmentioning
confidence: 99%
“…One alternative to using HPC machines is to use the cloud, however, this is not as attractive an option as it may initially seem. In a previous study for VESTEC [39], it was found that the performance of cloud systems tends to be significantly lower than a corresponding HPC machine, mainly due to filesystem performance, the cost was higher, and the cloud is not as elastic as one would hope, with reasonably long startup wait times for large, infrequent compute requirements. Whilst these make it impractical to rely upon a cloud-based resource, it would still be plausible to supplement the HPC machines we federate over with some cloud resources for some jobs/applications.…”
Section: Discussionmentioning
confidence: 99%
“…Whilst the computational power provided by supercomputers is key in enabling the timely simulations required for urgent workloads [5], such machines are often optimised for throughput rather than minimising the latency of individual jobs. This unbounded nature of the wait time in the queue means that in the past HPC machines are generally acknowledged as being less suited to running urgent workloads within specific time constraints.…”
Section: Vestec: An Ecosystem For Urgent Computingmentioning
confidence: 99%
“…During the COVID-19 pandemic, supercomputers and data systems have worked together with unprecedented efficiency, from understanding the structure of the SARS-CoV-2 virus to modeling its spread, from therapeutics to vaccines, from medical response to managing the virus's impacts [17,3]. Similarly, the VESTEC European FET project fuses super computers and real-time data for urgent decision-making [11].…”
Section: Introductionmentioning
confidence: 99%