Significant data reduction is essential for extreme-scale scientific facilities, where massive spatiotemporal datasets overwhelm traditional parallel file systems and create severe I/O bottlenecks. To mitigate this, emerging hybrid HPC workloads integrate Deep Neural Network (DNN) models into scientific workflows to perform predictive data compression. TEZip (Time Evolutionary Zip) uses these DNN-based methods to predict sequences and reduce storage footprints. However, integrating dynamic Python-based ML frameworks (e.g., PyTorch) into static HPC environments introduces severe runtime and data movement overheads, preventing these hybrid applications from keeping pace with node-local data generation rates. In this work, we present the first native C++/LibTorch implementation of TEZip, an architectural co-design explicitly targeting the HPC I/O critical path. We have updated the prediction module to use advanced deep learning architectures, including ConvLSTM and PredNet, to process spatiotemporal data. Rather than a simple language translation, we redesign the tensor lifecycle management and prediction loops to embed this inference directly into the I/O stream. We identify and quantify Python runtime overhead sources unique to these neural compression workloads, and expose non-trivial design challenges in embedding LibTorch inference into parallel I/O pipelines. This native C++ architecture achieves approximately a 4 × speedup in training, a 13 × speedup in compression, and a 4.8 × speedup in decompression in three datasets, while maintaining the identical compression ratio and reconstruction quality. These improvements allow AI-driven predictive compression to run efficiently at the edge of the compute tier, significantly reducing data volume before it affects parallel file system bandwidth.

TEZip is a framework for compressing time-evolving image data using predictive deep neural networks. Until now, TEZip primarily relied on Python libraries (TensorFlow or PyTorch). This work presents a new TEZip pipeline built with C++/LibTorch for improved speed and High preformance Computer(HPC) compatibility.

Abstract In recent years, large language models (LLMs) have gained significant traction across various domains, including education. This paper explores the application of LLMs in grading programming assignments. By leveraging data collected from existing programming assignments and their corresponding grades, we aim to develop a robust LLM-based grading system. We also incorporate augmented data representing various grading scenarios to enhance the model’s performance and ensure comprehensive coverage across all grading levels. Our approach involves training the LLM on this combined dataset to enable accurate and consistent evaluation of programming assignments. The proposed model, BeGrading, aims to reduce the grading burden on educators and provide timely and objective feedback to students. Compared to the Codestral model, our proposed model demonstrates an absolute difference rate of 19%, equivalent to $$\pm 0.95$$ ± 0.95 out of 5. This is acceptable for using a small, fine-tuned model with optimized data. Additionally, the Codestral model compared to the dataset optimized score shows a difference of 15% equivalent to a margin of $$\pm 0.75$$ ± 0.75 out of 5. Preliminary results demonstrate the potential of LLMs to perform grading tasks with a high degree of reliability, opening avenues for further research and practical applications in automated education systems.