2016/17 Vacation Scholarships



2016/17 Vacation ScholarshipsJob Title:CSIRO Undergraduate Vacation Scholarships – Information Management and TechnologyReference No:24365Classification:CSOF1.1 Stipend:$1462.77 per fortnight (before tax)Location:Please refer to the list of Projects at the end of this documentTenure:8 to 12 weeks from November 2016 to February 2017Role Purpose:The 2016/17 Vacation Scholarship Program is designed to provide students with the opportunity to work on real-world problems in a leading R&D organisation.Participation in the Vacation Scholarship Program has influenced previous scholarship holders in their choice of further study and future career options. Many have gone on to pursue a PhD in CSIRO or to build a successful research career within CSIRO, a university or industry.Project Description:Please refer to the list of Projects at the end of this document. If you require more information please contact the person listed for the project.Eligibility/ Pre-Requisites:To be eligible to apply you must be an Australian or New Zealand Citizen, Australian Permanent Resident or an international student who has full work rights for the 8 to 12 weeks duration (does not require visa sponsorship).Vacation scholarships are for students who:are currently enrolled at an Australian university; have completed at least three years of a full-time undergraduate course (however exceptional second year students may be considered);have a strong academic record (credit average or higher); andintend to go on to honours and/or postgraduate study.How to Apply: You will be required to:select your top 2 research projects in order of preference;submit a resume/cover letter (as one document) which includes:the reasons why the research project/s you have selected are of interest to you; and how your previous skills/knowledge and experience meets the project requirements; andan outline of your longer-term career aspirations and detail how this program will help you achieve them.upload your academic results in the ‘Requested Information’ field. Referees: If you would like to include referees (either work or university lecturers/ tutors) in your application, please add their name and contact details into your resume. If you experience difficulties applying online call 1300 984 220 and someone will be able to assist you. Outside business hours please email: csiro-careers@csiro.au. Please do not email your application. Applications received via this method may not be considered.Project Number– Information Management and Technology - Vacation Scholarships Project DetailsIM&T 1Project TitleIntroducing Computational Fluid Dynamics to High School Students.Project DescriptionTo prepare for the economy of the future, Australia is seeking an increase in the uptake of science, technology, engineering, and maths (STEM) subjects in schools. CSIRO and other institutions contribute to this through programs such as "ICT in Schools" and by hosting work experience students.The goal of this project is to build an application that could be deployed through such programs which demonstrates scientific computing to high school students.The application will explore computational fluid dynamics, with classroom students able to draw a 2D object and set air speed, and the software predicts the air flow around the object. At the core, the software will contain a real computational fluid dynamics engine. The vacation scholar would be responsible for selecting appropriate programming language, software architecture, and deployment model but would be well supported by IMT staff.Project Duties/TasksIntroduction to CSIRO Scientific ComputingIntroduction to ICT in Schools programLearn about Computation Fluid DynamicsSelect a technology stack and build a prototypeTest the software in a classroom environmentIterate on the software until suitableCreate a package for teachers / ICT professionals to teach CFD using the softwareStretch goalsImprove usability (user interaction)Extend the code to solve more realistic (complex) scientific computation.Relevant Fields of Study?Natural and physical sciences ?Information technology ?EngineeringAn awareness of Computational Fluid Dynamics (CFD) would be desirable but not essential.Location: Clayton, VICContact: Paulus Lahur via phone on (03) 9545 2382 or email paulus.lahur@csiro.au IM&T 2Project TitleMappingMadeEasy: an accessible web-based platform for interactive geospatial temporal visualisations.Project DescriptionVisualisations of spatial-temporal datasets facilitate the analysis of time and location varied events and identify different patterns in the data. CSIRO's scientific computing visualisation team are experiencing a rising number of requests to build geospatial temporal visualisations. Typically, geo-spatial datasets have variables associated to different geographic locations (specified by longitude, latitude) and multiple time points. Common visualisation techniques used for geospatial temporal datasets involve mapping the longitude and latitude locations for a single time point and animating through different time points with a time slider.The goal of this project is to evaluate common visualisation libraries (e.g. openlayers, leaflet, D3) and GIS tools/platforms (e.g. ArcGIS) and develop or deploy a web-based platform that enables the easy creation of interactive geo-spatial temporal visualisations for scientists.Project Duties/TasksEvaluate common visualisation libraries and GIS tools/platformsDevelop/set up a web-based platform that can create maps using point and polygon source geo-data.Implement the time slider and a basic brushing interactionRelevant Fields of StudyNatural and physical sciences Information technology EngineeringLocation: Clayton, VICContact: Florence Wang phone on (03) 95452378 or email Florence.Wang@csiro.auIM&T 3Project TitleAccessible Deep Learning For Every Scientist.Project DescriptionThe field of deep learning is one of the fastest growing and most exciting areas of computer science. The technology is already imbedded in our everyday lives, from Google search suggestions to online assistant bots, and soon we will even be driven to work by AI. While the underlying theory behind these developments is quite complex an intuitive understanding of the process is easy to obtain. This makes the field ripe for citizen scientists or curious researchers to make meaningful and impactful discoveries without years of technical training. Unfortunately the software suite available for deep learning is geared towards technical research in domain specific fields. With an interest in the hardware development, NVIDIA has attempted to overcome this hurdle by developing a GUI based, GPU enabled, application, DIGITS. This reduces the task of deep learning, in particular image classification, down to a series of button clicks. With such a low barrier to entry DIGITS would offer many scientists within CSIRO the opportunity to explore the benefits afforded by a deep learning solution. However, at present the DIGITS tool is difficult to use on our high powered GPU enabled computing cluster. It requires a technical knowledge beyond the otherwise low barrier to entry DIGITS provides and computing clusters are a poor match to interactive exploration. We need to develop a user friendly approach to launching, training, and classifying with DIGITS to give deep learning the opportunity to grow within CSIRO. Such a development would also offer the possibility of improved utilisation of our GPU cluster. Time permitting the new DIGITS implementation may be used to benchmark our current clusters for deep learning.Project Duties/TasksLearn / understand DIGITS and the deep learning concept.Integrate and batch launch option into the open source DIGITS code.Develop and extension to the SC Launcher which will enable scientists to interact with DIGITS without having any knowledge of the underlying batch system.Benchmark CSIROs GPU cluster using DIGITS.Relevant Fields of StudyNatural and physical sciences Information technology EngineeringLocation: Clayton, VICContact: Chris Watkins via on (03) 9545 2369 or email chris.watkins@csiro.auIM&T 4Project TitleWorkflow plug-in for CSIRO's Data Access Portal.Project DescriptionCSIRO has developed a workflow tool called Workspace and a research data repository called the Data Access Portal (DAP). A plug-in was written to make use of some of DAP’s REST web service capabilities, primarily for the purpose of data collection file retrieval.Since the DAP Workspace plug-in was first written, there have been improvements to the DAP web service API that have yet to be incorporated.This project involves extending the existing Workspace DAP plug-in to allow data collection creation as part of a workflow.Project Duties/TasksLearn about DAP and its REST web service.Learn about the Workspace workflow tool.Work through the Workspace plug-in development tutorial.Learn about version control with git and Bitbucket.Read & extend the C++ Workspace DAP plug-in source code to:Create and publish a draft collection.Transfer files to a draft collection’s staging area via REST and scp/sftp.Incorporate other useful web service functions, e.g. metadata retrieval, in consultation with supervisor and interested parties.Create simple Workspace workflows to test and demonstrate each plug-in feature.Create a Workspace workflow that implements an interesting scenario that makes use of the new DAP plug-in capabilities, in consultation with supervisor and interested parties.Initiate and take part in an online git-based code review.Write documentation for the new plug-in features.Relevant Fields of StudyNatural and physical sciences Information technology EngineeringPrevious exposure to C++ programming would be highly desirableLocation: Waite, SA (negotiable)Contact: David Benn via phone on (08) 8303 8512 or email David.Benn@csiro.au See also: 5Project TitleNational energy use data analysis, metadata extraction, and visualisation.Project DescriptionData is generated by people, systems and websites and is necessary for decision making and modelling. In the research environment we aim to produce data that is consistent, accurate and representative. For this to be achieved we need to understand the data, evaluate and address data quality and develop approaches to capture meaningful metadata. Metadata is defined as data about data and is needed to support the application of data for future re-use. In this project you will learn about the CSIRO scientific computing environment, become involved in the analysis of national energy use datasets, implement code to extract metadata and create high-resolution spatial temporal GIS maps. The application of these activities extends to most domains (Manufacturing, Water etc.) and is only dependant on data collection.Project Duties/TasksFamiliarisation with CSIRO scientific computing infrastructureProgramming using Python (Jupyter IDE)Store, analyse and curate dataImplement code to extract metadataFamiliarisation with ESRI ArcGISCreate high-resolution spatial temporal GIS maps to visualise energy dataRelevant Fields of StudyNatural and physical sciencesInformation technologyEngineeringLocation: Clayton, VICContact: Julia Anticev via phone on (03) 9545 2915 or email Julia.Anticev@csiro.auIM&T 6Project TitleVirtual Reality Interaction and Scientific Applications.Project DescriptionVirtual Reality (VR) continues to rise in the market as more technologies claim to be production-ready. The next two to three years is crucial to understand its user interaction in order to deliver a good design experience. This project seeks to investigate VR interaction paradigms. What is an easy task in VR? What is difficult? Why are some users confused, sitting still, and some other users move around much more? These are open questions in the current literature.Scientific Computing has several multivariate datasets that can be loaded into a virtual environment. For instance, geospatial datasets on CO2 concentrations in Australia; locations of bee flight paths recorded over time; blood vessels from a CT scan; so forth. Good interactions in VR with datasets like the ones above will accelerate scientific research using virtual tools.The goal is to produce a set of design recommendations driven by experimental prototypes. Tasks may be specific to fit scientific application needs, but they will be grounded on good design principles, the primary output of this project.Project Duties/TasksSynthesise the types of VR affordances in the literature and media.Carry out experiments to understand and simplify tasks. For example: orienting, picking up objects, fine finger/hand movements.Implement specific functions that fit scientific application needs.Produce a list of recommendation on good VR design principles.Relevant Fields of StudyDesign ComputingSoftware EngineeringComputer ScienceInformation TechnologyGame DesignLocation: Clayton, VICContact: Xavier Ho via email xavier.ho@csiro.au IM&T 7Project TitleFault Tolerant MPI.Project DescriptionMessage Passing Interface (MPI) is the most common Applications Programming Iinterface used to develop parallel applications for distributed memory systems used in technical and scientific computing.Most modern MPI libraries and applications are focused on performance and assume implicitly that the underlying hardware (nodes and interconnect) is reliable and hardware faults are rare and unlikely to occur.As the size of modern systems and simulations grows, long running applications are more likely to encounter hardware failures during their execution. Hence the importance of making the MPI libraries and applications fault tolerant to enable them to continue correct operation in the presence of hardware faults and other system errors.There are many approaches to fault-tolerance. This project focuses on the User Level Failure Mitigation (ULFM) approach developed by MPI Forum's Fault Tolerance Working Group. In this approach the MPI library is setup (at runtime) not to terminate the application in the presence of errors, but it is still the user's responsibility to develop application code that detects the errors and recovers from them in order to produce correct results despite system errors.Project Duties/TasksBuild and deploy a Fault Tolerant prototype of OpenMPI on a CSIRO's scientific computing infrastructure.Explore new MPI APIs from the ULFM proposal.Modify a simple, pre-existing MPI application to survive an abnormal termination of one or more application processes.Investigate approaches to continuous correct operation past abnormal termination of one or more application processes.Relevant Fields of StudyNatural and physical sciences Information technology EngineeringLocation: Clayton, VICContact: Maciej Golebiewski via phone on (03) 9545 2270 or email maciej.golebiewski@csiro.au IM&T 8Project TitleTree Canopy Segmentation with Apache Spark.Project DescriptionThe emergence of handheld laser scanners such as CSIRO's Zebedee make it easy for researchers to collect vast amounts of data in the form of point clouds. While tools exist to transform and visualise these point clouds, they generally operate on only a single computer which constrains the size of the datasets that can be used and limits processing speed.This project will use the popular and powerful Apache Spark framework to distribute large point cloud datasets across a cluster of computers, and then classify the points in the data as tree trunks or canopy.Project Duties/TasksFamiliarise with CSIRO's scientific computing infrastructureDeploy Apache Spark on CSIRO's internal cloudFamiliarise with Python and PySpark run from a Jupyter environmentUse spark to read in point cloud data from CSIRO's handheld laser scanner, the ZebedeeCreate a supervised learning implementation that coarsely separates tree canopies from other points in the datasetVisualise the segmented dataset to inspect accuracyAs an extension, the student could explore the use of the streaming machine learning algorithms in Spark to process data arriving in real-timeRelevant Fields of StudyNatural and physical sciences Information technology EngineeringLocation: Clayton, VICContact: Sam Moskwa via phone on (03) 9545 7845 or email sam.moskwa@csiro.au IM&T 9Project TitleHPC Containers (Shifter).Project DescriptionContainer-based computing allows software applications to run on almost any computer, independent of the host operating system. It does this by packaging the application with the entire software stack, including required system configurations, environment variables, and other dependencies. For scientific computing, these containers provide a lightweight means of encapsulating the lifecycle of a given compute task and support users in creating more reproducible and readable workflows. The aim of this project is to explore options for running containerised applications on CSIRO’s high performance computing (HPC) systems, complete a feasibility study, and where possible, deploy a prototype container system.Project Duties/TasksConfigure and deploy a instance of Shifter on a subset of a CSIRO compute cluster or virtual cluster.Configure Shifter to deploy several jobs using several known applications within IM&T (Python, R, and/or MATLAB)Evaluate the performance of the test system and of ShifterTime permitted, produce a report regarding the performance, pros/cons and complexities involved in deploying this type of technology in our environment.Relevant Fields of StudyNatural and physical sciences Information technology EngineeringLocation: Clayton, VICContact: Joel Ludbey via phone on (03) 9545 2024 or email Joel.Ludbey@csiro.au IM&T 10Project TitleSoftware Management/Automation.Project DescriptionHigh Performance Computing is used in every field of science. To obtain the best possible result, software is often compiled from source for the specific architecture with optimizations that are learnt and finely tuned over the years and which may change with new compilers, new software, and new hardware. These optimisations are currently performed manually on each system making them prone to human error and often require tedious repetition to fix compilation errors or bugs. This can introduce inconsistencies over time and undermines scientific reproducibility. One solution is to use a package management system that integrates with our scientific computing infrastructure, automating the process of building software across all systems. This would maintain a consistent software infrastructure from a single version controlled repository and be somewhat self documenting. Although numerous tools exist to do the job, they need to be evaluated and integrated with current systems and may involve modifying code upstream. Ideally, end users will require few or no changes in current behaviour with the new methods of building/installing and using software being a seamless backend change. Furthermore, automation of software management would save significant time, provide a better infrastructure for improving optimisation, allow continuous integration, and automated benchmarking capabilities in the future.Project Duties/TasksEvaluate and select one of multiple software management tools for use with existing CSIRO resources Investigate and proceed with integration of selected tool with existing resources, primarily for HPC use (modules, git {bitbucket}, jenkins) and systems (clusters, cloud) - this may require modifying upstream project source.If feasibly, implement desired solution as part of project.Relevant Fields of StudyNatural and physical sciences Information technology EngineeringLocation: St Lucia, QLDContact: Ondrej Hlinka via phone on (07) 3327 4106 or email ondrej.hlinka@csiro.au ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download