Weikuan Yu, Professor in the FSU Department of Computer Science, has been awarded a grant of $83K for research on I/O performance characterization and optimization from Lawrence Livermore National Lab (LLNL). The project, titled as “Analyzing and Optimizing I/O Performance of HPC”, conducts an extensive performance characterization for the VAST file system, which is a cutting-edge implementation of the disaggregated storage paradigm for productive deployment. The characterization will answer the question on the potential of VAST in addressing the scalability problems associated with large-scale I/O activities from the training of massive deep neural networks (DNNs) and large language models (LLMs). Built on top of insights from I/O characterizations, Dr. Yu and his team plan to optimize the I/O pipeline from these applications to the VAST file system.
