Session would include numerous papers which address exascale computing and preparations to use this next big leap in computing power.

Earth System Prediction relies heavily on High-Performance Computing (HPC); its needs are expanding as the demands of earth system modeling (ESM) push the limits of computational efficiency. Advances in private sector capabilities including Artificial Intelligence (AI) and cloud-based HPC may enhance the ability of academic and federal research communities to explore the constantly emerging frontiers of ESM.  Previous studies have shown the benefits and drawbacks of reliance upon proprietary systems. This session invites abstracts highlighting advances and challenges in HPC capability for research in ESM including, but not limited to, data storage, cloud computing, interconnect technology, code refactorization, computational efficiency, and evolving market-based priorities. The session seeks to incorporate industry perspectives on new architectures and capabilities, engagement with the ESM community and current ESM-related HPC.

Cloud computing is becoming pervasive in a broad range of computational domains and is being explored to determine the benefits to the Weather, Water and Climate community. The cloud promises lower cost, ease of use, ease of access, and flexible provisioning, but there are challenges to gaining these benefits and the cloud may not be appropriate for all workloads. This session solicits papers presenting experiences in using HPC in the cloud for Weather, Water and Climate applications as well as analyses addressing suitability of cloud platforms for delivering HPC to various Weather, Water and Climate applications. Speakers will address challenges and lessons learned from the experience, including issues of issues of performance, availability, data storage, data transfer and security.

HPC Poster Session

This session brings together the four conferences under the Board for Environmental Information Processing Technologies. Presentations should address the common technologies used across the weather enterprise. The views can be from the past, present or future. For example, what were the prevalent programming languages 30 years ago, 10 years ago, now and in the future.

This session will be a joint session between the 6th Symposium on High-Performance Computing for Weather, Water, and Climate (6HPC) and the 19th Conference on Artificial Intelligence for Environmental Science. As datasets grow larger and artificial intelligence methods grow more complex, researchers have pushed the need for more powerful computing resources to make scientific advances. This session will explore how these Big Data projects are enabled by researchers using new computing methods, greater levels of parallelism, faster storage systems, better optimized algorithms, and more to push artificial intelligence and machine learning forward. Example submissions may include (but are not limited to) using clusters of specialized hardware such as graphical processing units (GPUs) to enable shorter time to results or larger pools of results, optimized storage systems that can reduce the latency of delivering data to learning algorithms distributed over a high-performance system, or work creating or using advanced algorithms that improve scalability towards exascale computing.

This will be a joint session between the 6th Symposium on High-performance Computing for Weather, Water, and Climate (6HPC) and the 26th Conference on Numerical Weather Prediction (26NWP). This session will explore the use of high-performance computing to develop, run, enhance the performance of, and post-process output from numerical weather models such as WRF, convection-allowing models, convection-allowing ensembles, etc. Abstract topics may include (but are not limited to) improving computational efficiency, improving scalability, preparing for exascale computing, and optimizing numerical weather models to benefit from the use of specialized hardware in HPC environments, such as graphical processing units (GPUs).

This will be a joint session between the 6th Symposium on High-performance Computing for Weather, Water, and Climate (6HPC) and the 26th Conference on Numerical Weather Prediction (26NWP). This session will explore the use of high-performance computing to develop, run, enhance the performance of, and post-process output from numerical weather models such as WRF, convection-allowing models, convection-allowing ensembles, etc. Abstract topics may include (but are not limited to) improving computational efficiency, improving scalability, preparing for exascale computing, and optimizing numerical weather models to benefit from the use of specialized hardware in HPC environments, such as graphical processing units (GPUs).