BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
CALSCALE:GREGORIAN
PRODID:-//WordPress - MECv6.5.6//EN
X-ORIGINAL-URL:https://conference.eresearch.edu.au/
X-WR-CALNAME:
X-WR-CALDESC:eResearch Australasia Conference
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
X-MS-OLK-FORCEINSPECTOROPEN:TRUE
BEGIN:VEVENT
CLASS:PUBLIC
UID:MEC-fc32f8c132c0fb39c91d4f3dad7085a4@conference.eresearch.edu.au
DTSTART:20211012T010000Z
DTEND:20211012T012000Z
DTSTAMP:20210913T041900Z
CREATED:20210913
LAST-MODIFIED:20210913
PRIORITY:5
TRANSP:OPAQUE
SUMMARY:Transforming HPC Clusters with Modern Architecture and Configuration Management
DESCRIPTION:Dr Werner Scholz1, Jakub Szarlat\nXenon1\nCluster management tools and user interfaces are often the neglected layer between the users, their applications and the cluster hardware. But can a thoughtfully architected HPC cluster software stack enable modern cluster design, regardless of where the hardware lives?\nExploring the choice between cloud, on-premise or a hybrid infrastructure for HPC often revolves around comparisons of capabilities and costs – where critical factors are often unknown. This makes the decision one based on best assumptions of the time, with consequences that last years.\nThe real magic of modern HPC infrastructure at the cluster management layer frees one from vendor and/or cloud lock-in and allows for infrastructure flexibility, modular expansion. Resulting in an agile infrastructure that is composable and enables experimentation.\nJoin us for this thought-provoking Q&A session as Dr Werner Scholz and Jakub Szarlat from XENON push the boundaries of HPC infrastructure and explore the important role the cluster software stack plays for the user experience, management, and future proofing the whole system.\nHow can a modern software stack deliver flexibility, modular design, agile response to workloads?\nWhat is the best way to enable hybrid cloud workloads?\nCan on-premise clusters benefit from automation, scripting and an infrastructure-as-code approach?\nWe answer these questions and explore how key decision vectors such as dataset size and location, workloads and applications, bandwidth and security all come in to play.\n\nBiography:\nDr Werner Scholz, CTO and Head of R&D\nWerner has more than 15 years experience with high performance computing systems – from individual workstations and storage servers to massively parallel HPC clusters and large storage systems. He is also the developer of an open source finite element simulation package, which uses MPI, OpenMP, and GPU parallelization techniques. It is in use by academic and industrial research organizations around the world.\nWerner has a PhD in physics from the Vienna University of Technology in Austria, where he specialized in computational physics and magnetic materials. He is the author of more than 80 journal articles in the area of computational physics and magnetic nanostructures and co-inventor of 12 patents related to magnetic storage technologies.\nJakub Szarlat, Senior Systems Architect\nJakub is an experienced solution architect and infrastructure manager with a 20+ year history of working in the eResearch. Jakub has managed teams implementing and maintaining complex infrastructure environments in university and research environments. Jakub excels in realising innovative ideas, such as private clouds, containerisation in eResearch, and solutions to simplify management of complex VDI environments. Jakub’s approach to solution is to focus on helping scientists do science with the best IT available.\n
URL:https://conference.eresearch.edu.au/events/transforming-hpc-clusters-with-modern-architecture-and-configuration-management/
END:VEVENT
END:VCALENDAR