You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 18 Next »

Key research thrust areas:

There are three key thrust areas in QuakeCoRE's computational infrastructure efforts (The specifics for each of the thrust areas are outlined below in their respective pages):

  1. Ground motion simulation: Includes the development of a computational pathway for handling ground motion simulation calculations, the necessary input model information, post-processing of results and visualisation.  Further information can be found in this 2016 QuakeCoRE Annual Meeting Poster.

  2. Seismic response modelling of infrastructure: Includes both detailed modelling of individual infrastructure components (e.g. individual structural and geotechnical systems), as well as spatially-distributed infrastructure systems.
    • OpenSees Development: The Open System for Earthquake Engineering Simulation (OpenSees) is the principal collaborative software for detailed modelling of individual infrastructure used by QuakeCoRE.  You can find specific initiatives by clicking on the link above.
    • Further information on QuakeCoRE can be found in the 2016 QuakeCoRE Annual Meeting Poster for this topic.

  3. Seismic performance and loss assessment: Includes improving software for existing seismic performance and loss assessment tools for both regional and building-specific applications; connection of ground motion simulation and seismic response modelling of infrastructure computational pathways with loss assessment methods.  Further information on work to date can be seen in this 2016 QuakeCoRE Annual Meeting Poster.

 

Overarching objectives

The two key objectives for this platform are:

‘Reducing barriers to entry’: Simplifying the access to data and computational software workflows so that it is easier to use for researchers unfamiliar with specific nuances.  This will lead to an increase in the usage of high quality data and computational tools by QuakeCoRE researchers, both improving research quality and connectivity between disciplines.

‘Reducing time to solution’: Through optimizing database storage and computational codes, as well as pre- and post-processing workflows, we will allow researchers to speed up their research.  Improved timeliness will allow for the solution of both previously un-managable problems, as well as improved multi-disciplinary research interactions.

Principles

The underlying principles to attain the overarching objectives are:

Open Source: The software should be open source to maintain flexibility, enhance collaboration, and not be dependent on external software organisations.  This also recognises that NZ researchers represent a small portion of the global human resource in this field and the use of OS software enables a greater leverage of international initiatives.

Scalable: The software should be able to be scaled to make use of HPC facilities, particularly through NeSI.

Flexible: The outputs of the different software modules should be able to easily read by other software modules within the QuakeCoRE ecosystem.

Personnel

The QuakeCore (and aligned) staff for Technology Platform 4 (Simulation and Data Visualization)  (some of who are also involved in TP 3 (Multi-disciplinary community datasets)) are:

Sung Bae, IT architect

Sharmila Savarimuthu, software engineer

Viktor Polak, software engineer

Richard Clare (QuakeCoRE employee till June 2016)

Daniel Lagrava (Aligned software developer)

Michael Gauland (Aligned software developer)

The recent, on-going, and future activities tasks for  each of the three can be accessed by following the link on their name.

 

Key performance indicators:

  • TP4.1: Implement emod3d and specfem ground motion simulation software on NeSI HPC resources (by 1/1/2017)

  • TP4.2: Develop a streamlined workflow (incl. visualization) for ground motion simulation calculations to minimize the barrier to entry in new users running a large number of ground motion (by 1/1/2018)

  • TP4.3: Develop automated simulation of ground motions from NZ earthquakes in real-time (by 1/7/2018)

  • TP4.4: Optimize OpenSees FE software on NeSI HPC resources and train the QuakeCoRE community to develop a critical mass of capable OpenSees analysts (by 1/1/2018)

  • TP4.5: Develop centralized pre- and post-processing functionality for OpenSees calculations to reduce barriers to users to streamline their analyses (by 1/1/18)

  • TP4.6: Examine the parallelization of the OpenSees source code and look to improve its HPC scalability to enable large computational models to be run efficiently (by 1/1/2019)

  • TP4.7: Implement structure-specific and regional loss estimation software into a computational workflow with ground motion and OpenSees simulations (by 31/12/2020)

  • TP4.8: Implement multi-scale loss estimation methods which combine the details of current structure-specific methodologies at the regional level (by 31/12/2020)

  • TP4.9: Parallelize loss estimation software so that multi-scale loss estimation methods are computational feasible (by 31/12/2020)

     

 

 

  • No labels