BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:SIGHPC Edu DTSTART;VALUE=DATE-TIME:20241203T125500Z DTEND;VALUE=DATE-TIME:20241203T130000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2518@events.chpc.ac.za DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2518/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2518/ END:VEVENT BEGIN:VEVENT SUMMARY:® SIGHPC Committee DTSTART;VALUE=DATE-TIME:20241203T124500Z DTEND;VALUE=DATE-TIME:20241203T125500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2517@events.chpc.ac.za DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2517/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2517/ END:VEVENT BEGIN:VEVENT SUMMARY:® Onboarding new first time HPC users DTSTART;VALUE=DATE-TIME:20241203T123500Z DTEND;VALUE=DATE-TIME:20241203T124500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2516@events.chpc.ac.za DESCRIPTION:Speakers: John Poole ()\nhttps://events.chpc.ac.za/event/139/c ontributions/2516/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2516/ END:VEVENT BEGIN:VEVENT SUMMARY:Are You Doing Science? DTSTART;VALUE=DATE-TIME:20241203T115500Z DTEND;VALUE=DATE-TIME:20241203T123500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2515@events.chpc.ac.za DESCRIPTION:Speakers: Jay Lofstead (Sandia National Laboratories)\nhttps:/ /events.chpc.ac.za/event/139/contributions/2515/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2515/ END:VEVENT BEGIN:VEVENT SUMMARY:Overview of new local HPC course DTSTART;VALUE=DATE-TIME:20241203T114500Z DTEND;VALUE=DATE-TIME:20241203T115500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2514@events.chpc.ac.za DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2514/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2514/ END:VEVENT BEGIN:VEVENT SUMMARY:Overview of Wits HPC DTSTART;VALUE=DATE-TIME:20241203T113500Z DTEND;VALUE=DATE-TIME:20241203T114500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2513@events.chpc.ac.za DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2513/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2513/ END:VEVENT BEGIN:VEVENT SUMMARY:Welcome and Introduction DTSTART;VALUE=DATE-TIME:20241203T113000Z DTEND;VALUE=DATE-TIME:20241203T113500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2512@events.chpc.ac.za DESCRIPTION:Speakers: Bryan Johnston (CHPC)\nhttps://events.chpc.ac.za/eve nt/139/contributions/2512/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2512/ END:VEVENT BEGIN:VEVENT SUMMARY:HPC Ecosystems BoF DTSTART;VALUE=DATE-TIME:20241204T113000Z DTEND;VALUE=DATE-TIME:20241204T130000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2510@events.chpc.ac.za DESCRIPTION:Speakers: Coordinator: Mr Bryan Johnston (CHPC)\nThis session is an opportunity for members of the HPC Ecosystems Community and those th at identify as associates / affiliates to convene in-person. The session w ill allow for members to discuss matters relating to HPC Ecosystems Projec t as well as broader African HPC and emerging HPC community topics. The 90 -minute session will include 60-minutes of prepared talks from members of the community\, followed by a further 30-minutes of open time for discussi on and meaningful community engagement. Alas\, muffins are not guaranteed. \n\nProvisional talks include:\n\n* HPC Ecosystems Project review for 2024 \n* University of the Witwatersrand - Open OnDemand regional Lab\n* HPC Ec osystems monitoring toolchain\n* Wits SIG HPC with Ecosystems\n* A curricu lum for SysAdmin training\n* Feedback on an African HPC discovery survey\n * HPC Ecosystems Project plans for 2025+\n\nhttps://events.chpc.ac.za/even t/139/contributions/2510/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2510/ END:VEVENT BEGIN:VEVENT SUMMARY:Sustainable AI Infrastructure: A Mandatory Pathway for Supercomput ing Centres DTSTART;VALUE=DATE-TIME:20241204T094000Z DTEND;VALUE=DATE-TIME:20241204T102000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2506@events.chpc.ac.za DESCRIPTION:Speakers: Maximilian Höb (Leibniz Supercomputing Centre (LRZ) )\nAs artificial intelligence drives more and more innovation in industry and science with the emergence of Large-Language Models (LLMs)\, the deman d for such computing resources is growing substantially. This development poses major challenges for supercomputing centres around the world\, as th ey need to provide state-of-the-art AI resources while meeting their susta inability goals and mandates.\nWhile today's world-leading supercomputer a ccording to the Top500 list require about 23 MWatt (Frontier) to provide c omputing power to the scientific community\, we see an exponential increas e in power consumption for AI training and inference. Estimates for traini ng and interference of current LLMs are huge\, and newer models will requi re even more power\, representing an unnoticed increase in energy consumpt ion.\nFor the operational requirements of HPC and AI supercomputers\, we n eed to evaluate the total power consumption and thus the total energy base d on the Power Usage Effectiveness (PUE). A four-pillar model was develope d at the Leibniz Supercomputing Centre (LRZ) to implement a holistic optim isation strategy for energy efficiency\, which includes the building infra structure\, system hardware and software as well as the actual application s. By extending this to AI hardware and cooling\, optimising AI models and implementing AI-driven resource management\, supercomputing facilities ca n meet these requirements while offsetting their environmental footprint.\ n\nhttps://events.chpc.ac.za/event/139/contributions/2506/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2506/ END:VEVENT BEGIN:VEVENT SUMMARY:Empowering Southern Africa's Researchers: Overcoming HPC Barriers with MATLAB Tools DTSTART;VALUE=DATE-TIME:20241204T090000Z DTEND;VALUE=DATE-TIME:20241204T094000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2504@events.chpc.ac.za DESCRIPTION:Speakers: Marco Rossi (MathWorks)\nAs scientific data gets big ger and its analyses involves increasingly complex algorithms\, access to High Performance Computing (HPC) resources in the form of multi-core CPUs and GPUs become essential for scientific research.\n\nOne main impediment to harnessing these resources remains the high barrier to entry – HPC sy stems often require a deep understanding of computing architecture and com mand line interfaces\, a skill not many researchers possess.\n\nThe approa ch to parallel computing within MATLAB is to abstract away a lot of this c omplexity from the end-user\, so that they can progress seamlessly from th eir prototype on the desktop onto the cloud or the cluster. The power of t he [Parallel Computing Toolbox][1] and [MATLAB Parallel Server][2] is free and available already to more than 20+ Universities and Institutions in S outhern Africa\, [like CSIR and many others][3]. In this presentation\, I will show how MathWorks tools can streamline research computational effort with three ways to make parallel programming more accessible\n\n 1. the a bility to scale from local machine to cluster using minimal changes to the code\,\n 2. the ability to submit jobs remotely to HPC clusters from with in the users’ local MATLAB environment using MATLAB Parallel Server\,\n 3. free training workshop being held globally in partnership with HPC cent ers to upskill scientists and researchers in parallel programming.\n\n\n [1]: https://nl.mathworks.com/products/parallel-computing.html\n [2]: htt ps://nl.mathworks.com/products/matlab-parallel-server.html\n [3]: https:/ /nl.mathworks.com/academia/tah-portal/csir-31652745.html\n\nhttps://events .chpc.ac.za/event/139/contributions/2504/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2504/ END:VEVENT BEGIN:VEVENT SUMMARY:AI next trends: the critical HW and SW marriage (Intel) DTSTART;VALUE=DATE-TIME:20241202T093500Z DTEND;VALUE=DATE-TIME:20241202T095500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2456@events.chpc.ac.za DESCRIPTION:Speakers: Walter Riviera (Intel)\nIn this session we'll walk t hrough the rapid evolution of the AI workloads\, and learn key insights on how next gen Hardware and Software will look like.\n\nhttps://events.chpc .ac.za/event/139/contributions/2456/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2456/ END:VEVENT BEGIN:VEVENT SUMMARY:Digital Twins for Energy Efficient and Sustainable Datacenters (HP E) DTSTART;VALUE=DATE-TIME:20241202T091500Z DTEND;VALUE=DATE-TIME:20241202T093500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2455@events.chpc.ac.za DESCRIPTION:Speakers: Tim Dykes ()\nDigital Twins are the basis of a rapid ly growing field of study that pairs physical systems with their digital r epresentation to enhance understanding of the physical counterpart. Digita l twins of data-centres are becoming a reality\, using bi-directional feed back loops to link operational telemetry with models of associated sub-sys tems\, which are combined with visualisation and analytics components. We will discuss on-going collaborative research into such datacenter digital twins\, which promise improved system behaviour prediction\, optimised sys tem operation\, and ultimately better-informed decision-making for operati ng large-scale compute centre installations with associated power\, coolin g\, and management infrastructure in a sustainable and energy efficient ma nner. \n \nTim Dykes | Senior Research Engineer | HPE HPC/AI EMEA Research Lab\n\nhttps://events.chpc.ac.za/event/139/contributions/2455/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2455/ END:VEVENT BEGIN:VEVENT SUMMARY:The Data Platform to Accelerate HPC Deployments (VAST Data) DTSTART;VALUE=DATE-TIME:20241202T121000Z DTEND;VALUE=DATE-TIME:20241202T123000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2453@events.chpc.ac.za DESCRIPTION:Speakers: Kim Ormiston (Sponsor)\nThe VAST Data Platform accel erates HPC by delivering the performance and scale of parallel file system s with the simplicity of NAS\, all at archive economics. Powered by VAST ’s DASE architecture\, our customers build scale-out systems delivering TB/s and millions of IOPS of performance without the rigidity and overhead of legacy architectures.\n\nhttps://events.chpc.ac.za/event/139/contribut ions/2453/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2453/ END:VEVENT BEGIN:VEVENT SUMMARY:NVIDIA Accelerated Computing (NVIDIA) DTSTART;VALUE=DATE-TIME:20241202T123000Z DTEND;VALUE=DATE-TIME:20241202T125000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2454@events.chpc.ac.za DESCRIPTION:Speakers: Belinda Leigh Duke (Partner)\nNVIDIA’s Accelerated Computing Platform is designed to accelerate various workloads\, includin g AI\, HPC\, Professional Visualisation & Gaming. The platform provides fa ster performance\, increased productivity\, improved accuracy\, and cost s avings for organizations across industries.\n \nNVIDIA’s Accelerated Com puting Platform is designed to accelerate various workloads\, including:\n \n* Artificial Intelligence (AI): NVIDIA’s GPUs and Tensor Cores accele rate AI workloads\, such as deep learning\, natural language processing an d computer vision.\n* High Performance Computing (HPC): NVIDIA’s GPUs an d networking technologies accelerate HPC workloads\, such as scientific si mulations\, data analytics and visualization\n* Professional Visualization : NVIDIA’s GPUs and Quadro professional visualisation workloads\, such a s 3D modelling\, engineering\, and video production\n* Gaming: NVIDIA’S GeForce GPUs accelerate gaming workloads\, providing fast performance\, lo w latency\, and high-quality graphics\n \nNVIDIA’s Accelerated Computing Platform:\n \n* NVIDIA GPU Architecture: NVIDIA’s GPU Architecture is d esigned to accelerate compute -intensive workloads\n* CUDA: CUDA is NVIDIA ’s parallel computing platform and programming model and allows develope rs to harness the power of NVIDIA GPU’S\n* NVIDIA Networking: NVIDIA’s networking technologies such as NVIDIA Mellanox\, provide high-speed netw orking solutions for accelerated computing workloads\n* NVIDIA Software: N VIDIA provides a range of software development kits (SDKs) and tools to su pport accelerated computing workloads.\n \nNumerous Industries benefit fro m NVIDIA Accelerated Computing\, just to name a few: Healthcare (accelerat ion of medical imaging\, genomics\, and drug discovery workloads) Finance (acceleration of financial modelling\, rick analysis and trading workloads ) Scientific Research: (acceleration of scientific simulations\, data anal ytics and visualization workloads) Gaming and Professional Visualization: (faster performance and high-quality graphics)\n\nhttps://events.chpc.ac.z a/event/139/contributions/2454/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2454/ END:VEVENT BEGIN:VEVENT SUMMARY:Q&A DTSTART;VALUE=DATE-TIME:20241202T103500Z DTEND;VALUE=DATE-TIME:20241202T104500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2491@events.chpc.ac.za DESCRIPTION:Speakers: Q&A ()\nhttps://events.chpc.ac.za/event/139/contribu tions/2491/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2491/ END:VEVENT BEGIN:VEVENT SUMMARY:Pinnacle Presentation (TBC) DTSTART;VALUE=DATE-TIME:20241202T101500Z DTEND;VALUE=DATE-TIME:20241202T103500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2458@events.chpc.ac.za DESCRIPTION:Speakers: Roger Crighton (Supermicro)\nTBC\n\nhttps://events.c hpc.ac.za/event/139/contributions/2458/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2458/ END:VEVENT BEGIN:VEVENT SUMMARY:Dell AI Factory (Dell) DTSTART;VALUE=DATE-TIME:20241202T095500Z DTEND;VALUE=DATE-TIME:20241202T101500Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2457@events.chpc.ac.za DESCRIPTION:Speakers: Jacques Visagie ()\nTBC\n\nhttps://events.chpc.ac.za /event/139/contributions/2457/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2457/ END:VEVENT BEGIN:VEVENT SUMMARY:Q&A DTSTART;VALUE=DATE-TIME:20241203T102000Z DTEND;VALUE=DATE-TIME:20241203T103000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2484@events.chpc.ac.za DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2484/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2484/ END:VEVENT BEGIN:VEVENT SUMMARY:Q&A DTSTART;VALUE=DATE-TIME:20241203T145000Z DTEND;VALUE=DATE-TIME:20241203T150000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2477@events.chpc.ac.za DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2477/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2477/ END:VEVENT BEGIN:VEVENT SUMMARY:Q&A DTSTART;VALUE=DATE-TIME:20241202T125000Z DTEND;VALUE=DATE-TIME:20241202T130000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2463@events.chpc.ac.za DESCRIPTION:Speakers: Q&A ()\nQ&A\n\nhttps://events.chpc.ac.za/event/139/c ontributions/2463/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2463/ END:VEVENT BEGIN:VEVENT SUMMARY:Huawei Presentation (TBC) DTSTART;VALUE=DATE-TIME:20241202T115000Z DTEND;VALUE=DATE-TIME:20241202T121000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2452@events.chpc.ac.za DESCRIPTION:Speakers: Collen Maphosa ()\nTBC\n\nhttps://events.chpc.ac.za/ event/139/contributions/2452/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2452/ END:VEVENT BEGIN:VEVENT SUMMARY:H3C Presentation (TBC) DTSTART;VALUE=DATE-TIME:20241202T113000Z DTEND;VALUE=DATE-TIME:20241202T115000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2451@events.chpc.ac.za DESCRIPTION:Speakers: Christo Cilliers ()\nTBC\n\nhttps://events.chpc.ac.z a/event/139/contributions/2451/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2451/ END:VEVENT BEGIN:VEVENT SUMMARY:Infrastructure consideration for Science Engagement Information Ma nagement Systems on CHPC DTSTART;VALUE=DATE-TIME:20241203T100000Z DTEND;VALUE=DATE-TIME:20241203T102000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2443@events.chpc.ac.za DESCRIPTION:Speakers: Muzi Matyila* (CSIR)\nThe Science Engagement Informa tion Management Systems (SEIMS) is a digitalisation initiative of the Depa rtment of Science and Innovation (DSI)’s Science Engagement (SE) Program me in collaboration with the CSIR-NGEI and other stakeholders. The SE prog ramme has the main objective of creating a society that is scientifically literate towards socio-economic emancipation. SEIMS consists of a number o f initiatives run by different stakeholders with the aim of expediting the monitoring and evaluation processes in SE programmes by eradicating manua l data management. The infrastructure of SEIMS is split between production and testing environments. The production environment is publicly availabl e and requires high availability. As a result\, a topology that supports t hese requirements was adopted. On the other hand\, the testing environment mainly required availability and a topology supporting this requirement h ad to be adopted. For this purpose\, the CHPC infrastructure\, a DSI suppo rted cyber-infrastructure was utilised. The CHPC\ncollaborated with the CS IR-NGEI’s Software Architecture and Solutions (SAS) group for the sourci ng and configuration of the infrastructure necessary for the SEIMS initiat ive. This presentation with outline experiences and lessons learned throug h utilization of the infrastructure and how it helped fulfill the requirem ents and demands of the SEIMS product.\n\nhttps://events.chpc.ac.za/event/ 139/contributions/2443/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2443/ END:VEVENT BEGIN:VEVENT SUMMARY:Bridging the Digital Divide with Cloud-Hosted Geo-Location Spectru m Databases DTSTART;VALUE=DATE-TIME:20241203T090000Z DTEND;VALUE=DATE-TIME:20241203T092000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2442@events.chpc.ac.za DESCRIPTION:Speakers: Luzango Mfupe* (CSIR)\nAffordable and widespread bro adband access is essential for achieving the United Nations Sustainable De velopment Goals (UN SDGs) cantered on connectivity. This presentation high lights efforts to close the digital gap in South Africa’s underserved ru ral and township communities through innovative use of a CSIR-developed\, Geo-Location Spectrum Database technology\, hosted by the Centre for High Performance Computing. Utilizing Dynamic Spectrum Access principles\, this technology taps into underused radio spectrum to enable wireless network operators to deliver cost-effective wireless broadband to remote areas. Th is CSIR developed technology has empowered women- and youth-led ICT SMMEs to deploy and manage low-cost network infrastructure\, providing affordabl e internet to schools\, businesses\, and households in previously disadvan taged regions. By creating job opportunities and fostering digital inclusi on\, this project supports national goals to accelerate progress towards t he UN SDGs.\n\nhttps://events.chpc.ac.za/event/139/contributions/2442/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2442/ END:VEVENT BEGIN:VEVENT SUMMARY:OpenStack Adoption at CHPC: Advancing AFIS and Emerging DevOps Pro jects Through Scalable Cloud Solutions DTSTART;VALUE=DATE-TIME:20241203T094000Z DTEND;VALUE=DATE-TIME:20241203T100000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2438@events.chpc.ac.za DESCRIPTION:Speakers: Kirodh Boodhraj* (CSIR)\nOpenStack is a cloud comput ing platform utilized by the Centre for High Performance Computing (CHPC) to manage a wide range of software services. Its user-friendly interface e nables the creation\, scaling\, and management of virtual machines with mi nimal effort\, offering a flexible and scalable Infrastructure as a Servic e (IaaS) solution. The Advanced Fire Information System (AFIS)\, is a sate llite-based tool that delivers near real-time fire information to users gl obally. AFIS is an internationally acclaimed Software as a Service (SaaS) which is traditionally hosted on self-managed infrastructure. AFIS’s bac kend and frontend systems are being transitioned to CHPC's OpenStack envir onment. This migration has been successful\, with several AFIS subsystems deployed without performance issues. The shift to OpenStack provides signi ficant advantages\, such as mitigating the risks associated with infrastru cture failure and reducing the need for manual intervention in server mana gement. Costs associated with server maintenance and infrastructure are si gnificantly saved on. Minimal modifications were required for the AFIS sub systems software deployment. Looking ahead\, OpenStack presents a promisin g solution for migrating other current projects\, including community safe ty project backends and geocoding engines and emerging DevOps applications such as the TB\, E-Coli and HPV projects. Lastly\, the integration of Ope nStack has facilitated enhanced collaboration with other CSIR groups\, CHP C\, SANReN\, and DIRISA.\n\n**Team members:** Kirodh Boodhraj (Project Lea d)\, Tau Makgaile\, Karen Steenkamp\, Gert Wessels\, Chris Mahlathi\, Phel isa Ntayiya\, Roger Daniels and Lufuno Vhengani (previous team member).\n\ nhttps://events.chpc.ac.za/event/139/contributions/2438/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2438/ END:VEVENT BEGIN:VEVENT SUMMARY:Image based analysis and crop prediction using smartphone. DTSTART;VALUE=DATE-TIME:20241203T092000Z DTEND;VALUE=DATE-TIME:20241203T094000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2437@events.chpc.ac.za DESCRIPTION:Speakers: Lindokuhle Mdletshe* (INNOBIZ Group (Pty) Ltd)\nThe 2019 White Paper on Science\, Technology\, and Innovation in South Africa\ nemphasises the core themes of inclusivity\, transformation\, and partners hips. Following\nthe introductory discussion with the farmers\, it became clear that the majority of them have\nan urgent need for up-to-date soil-h ealth data and assistance with managing their crops\nin order to maximize their profits. Addressing this\, our consortium aims to create a\nsmartpho ne app that will empower small-scale farmers with immediate soil-health\np redictions derived from soil digital images. Additionally\, this app will extend its\nfunctionality to provide crop recommendations and assist the f armer in smart decision-\nmaking by incorporating regular weather updates and nowcasting of local weather\nparameters. An integrated energy and sust ainability management module will be\ndesigned to monitor and control ener gy consumption and carbon emission. This project\nproposal unites partners from South Africa\, Scotland\, and Turkey\, leveraging their\ncollective expertise in science\, agriculture and digital technologies.\n\n\nThis inn ovation integrates advanced technologies such as digital image processing (DIP)\,\nmachine learning (ML)\, and artificial intelligence (AI)\, with d eep-learning algorithms and\nlarge language models (LLM) used for ML train ing\, and AI chatbot services for enhanced\nfunctionality. The app archite cture includes a mobile front-end API connected to a\ndynamic back-end sys tem\, featuring a database\, knowledge base\, and external APIs\, with Pyt hon serving as the core programming language. Therefore\, this project is using\nSebowa Cloud Services – Commercial (4x vCPU\, 16GB RAM\, 200 GB D isk).\n\nhttps://events.chpc.ac.za/event/139/contributions/2437/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2437/ END:VEVENT BEGIN:VEVENT SUMMARY:® I/O Insights and Recommendations for All DTSTART;VALUE=DATE-TIME:20241203T141000Z DTEND;VALUE=DATE-TIME:20241203T143000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2424@events.chpc.ac.za DESCRIPTION:Speakers: Jean Luca Bez (Lawrence Berkeley National Laboratory )\nThe complexity of the HPC I/O stack combined with gaps in the state-of- the-art profiling tools creates a barrier that does not help end-users and scientific application developers solve the I/O performance problems they encounter. To bridge this gap\, we introduce Dristhi\, a multi-source int eractive analysis framework that empowers users to visualize I/O traces\, identify bottlenecks\, and gain a deeper understanding of application beha vior. By combining cross-layer analysis with heuristic-based automatic det ection\, Dristhi provides actionable insights and recommendations to overc ome common I/O performance bottlenecks. This talk will delve into the desi gn and capabilities of Dristhi\, demonstrate its use in pinpointing I/O pe rformance issues\, and highlight upcoming features that will further enhan ce its functionality.\n\nhttps://events.chpc.ac.za/event/139/contributions /2424/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2424/ END:VEVENT BEGIN:VEVENT SUMMARY:® XIO: Towards eXplainable I/O using multi-layer monitoring solut ion for HPC Systems. DTSTART;VALUE=DATE-TIME:20241203T143000Z DTEND;VALUE=DATE-TIME:20241203T145000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2413@events.chpc.ac.za DESCRIPTION:Speakers: Hari Hariharan Devarajan (LLNL)\nModern HPC workload s exchange vast amounts of data to drive scientific discoveries While HPC systems employ diverse storage devices and tiers to support efficient data access\, current monitoring infrastructures\, such as Darshan and Score-P \, only provide enough information to show what they see\, but lack the vi sibility needed to fully explain observed I/O performance.\n \nIn this tal k\, I will present our latest survey of state-of-the-art monitoring tools deployed on modern HPC system using lists such as TOP500\, Green500\, IO50 0\, and the Comprehensive Data Center List (CDCL). Then\, we will introduc e our latest efforts to tackle the opaque monitoring infrastructure that e xplores the user and kernel I/O stack to uncover causality relationships f or achieving eXplainable I/O (XIO).\n\nhttps://events.chpc.ac.za/event/139 /contributions/2413/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2413/ END:VEVENT BEGIN:VEVENT SUMMARY:Characterizing the I/O Performance of HPC Platforms Using the I/O Performance Evaluation Suite DTSTART;VALUE=DATE-TIME:20241203T135000Z DTEND;VALUE=DATE-TIME:20241203T141000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2412@events.chpc.ac.za DESCRIPTION:Speakers: Luan Teylo* (INRIA Bordeaux)\nThe I/O Performance Ev aluation Suite (IOPS)\, currently under development by the TADaaM team at INRIA Bordeaux\, aims to streamline the performance analysis of I/O system s on HPC platforms. The primary objective of this tool is to automate the performance evaluation process\, initially determining the optimal combina tion of parameters (such as the number of nodes\, processes\, file sizes\, and access patterns) that enable the system to achieve peak I/O performan ce. Moreover\, IOPS can also be used to examine how parameters of parallel file systems\, such as the number of Object Storage Targets\, affect syst em performance under distinct scenarios.\n\nhttps://events.chpc.ac.za/even t/139/contributions/2412/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2412/ END:VEVENT BEGIN:VEVENT SUMMARY:Reducing Data Movement Through Derived Quantity Generation DTSTART;VALUE=DATE-TIME:20241203T133000Z DTEND;VALUE=DATE-TIME:20241203T135000Z DTSTAMP;VALUE=DATE-TIME:20260310T183652Z UID:indico-contribution-653-2411@events.chpc.ac.za DESCRIPTION:Speakers: Jay Lofstead* (Sandia National Laboratories)\nLearni ng from scientific simulations often relies not on the raw quantities calc ulated\, but instead from derived values. For example\, air pressures in a weather simulation in isolation are not as interesting as the air pressur e gradient. If the gradients are roughly parallel\, you have straight-line winds. If the gradient is rotating around a point\, it is a cyclone storm . Other uses\, such as reducing from a complex multi-quantity 3D model int o a 2D representation for the actual value required for analysis can radic ally reduce data movement from the large computational system to the local analysis system. We have built tools that enable automatically calculatin g these derived quantities as part of normal simulation output to enable e xtracting the derived quantities for movement to a different machine and o ther purposes.\n\nhttps://events.chpc.ac.za/event/139/contributions/2411/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2411/ END:VEVENT END:VCALENDAR