More Like this

1. TC-GNN: Bridging Sparse GNN Computation and Dense Tensor Cores on GPUs

   Wang, Yuke; Feng, Boyuan; Wang, Zheng; Huang, Guyue; Ding, Yufei. (July 2023, USENIX Association)

   Recently, graph neural networks (GNNs), as the backbone of graph-based machine learning, demonstrate great success in various domains (e.g., e-commerce). However, the performance of GNNs is usually unsatisfactory due to the highly sparse and irregular graph-based operations. To this end, we propose TC-GNN, the first GNN acceleration framework based on GPU Tensor Core Units (TCUs). The core idea is to reconcile the "Sparse" GNN computation with the high-performance "Dense" TCUs. Specifically, we conduct an in-depth analysis of the sparse operations in mainstream GNN computing frameworks. We introduce a novel sparse graph translation technique to facilitate TCU processing of the sparse GNN workload. We implement an effective CUDA core and TCU collaboration design to fully utilize GPU resources. We integrate MGG with the PyTorch framework for high programmability. Rigorous experiments show an average of 1.70× speedup over the state-of-the-art DGL framework across various models and datasets. 

   more » « less
2. Deep Learning Compilers

   Khatchadourian, Raffi (May 2026, CUNY Academic Works (https://academicworks.cuny.edu/gc_oers/16))

   These lecture slides introduce deep learning compilers for a graduate compiler-construction course (CSc 81010). Building on the classical compiler pipeline, they show how modern machine-learning systems compile tensor programs: static tensor and type analysis (illustrated by a WALA/Ariadne-based refactoring of imperative TensorFlow code to graph mode), MLIR-based end-to-end compilation with IREE, and the PyTorch 2.x stack—TorchDynamo graph capture, AOTAutograd, PrimTorch operator decomposition, and TorchInductor lowering to Triton (GPU) and C++/OpenMP (CPU). 

   more » « less
3. Towards Safe Automated Refactoring of Imperative Deep Learning Programs to Graph Execution

   https://doi.org/10.1109/ASE56229.2023.00187  

   Khatchadourian, Raffi; Castro-Vélez, Tatiana; Bagherzadeh, Mehdi; Jia, Nan; Raja, Anita (September 2023, IEEE/ACM International Conference on Automated Software Engineering)

   Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code-supporting symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, imperative DL frameworks encouraging eager execution have emerged at the expense of run-time performance. Though hybrid approaches aim for the “best of both worlds,” using them effectively requires subtle considerations to make code amenable to safe, accurate, and efficient graph execution. We present our ongoing work on automated refactoring that assists developers in specifying whether and how their otherwise eagerly-executed imperative DL code could be reliably and efficiently executed as graphs while preserving semantics. The approach, based on a novel imperative tensor analysis, will automatically determine when it is safe and potentially advantageous to migrate imperative DL code to graph execution and modify decorator parameters or eagerly executing code already running as graphs. The approach is being implemented as a PyDev Eclipse IDE plug-in and uses the WALA Ariadne analysis framework. We discuss our ongoing work towards optimizing imperative DL code to its full potential. 

   more » « less
4. Deep RC: A Scalable Data Engineering and Deep Learning Pipeline

   Sarker, Arup; Alsaadi, Aymen; Halpern, Alexander; Tangella1, Prabhath; Titov, Mikhail; Perera, Niranda; Staylor, Mills; von_Laszewski, Gregor; Jha, Shantenu; Fox, Geoffrey (June 2025, 28th edition of the workshop on Job Scheduling Strategies for Parallel Processing. JSSPP 2025 https://jsspp.org/)

   Significant obstacles exist in scientific domains including genetics, climate modeling, and astronomy due to the management, preprocess, and training on complicated data for deep learning. Even while several large-scale solutions offer distributed execution environments, open-source alternatives that integrate scalable runtime tools, deep learning and data frameworks on high-performance computing platforms remain crucial for accessibility and flexibility. In this paper, we introduce Deep Radical-Cylon(RC), a heterogeneous runtime system that combines data engineering, deep learning frameworks, and workflow engines across several HPC environments, including cloud and supercomputing infrastructures. Deep RC supports heterogeneous systems with accelerators, allows the usage of communication libraries like MPI, GLOO and NCCL across multi-node setups, and facilitates parallel and distributed deep learning pipelines by utilizing Radical Pilot as a task execution framework. By attaining an end-to-end pipeline including preprocessing, model training, and postprocessing with 11 neural forecasting models (PyTorch) and hydrology models (TensorFlow) under identical resource conditions, the system reduces 3.28 and 75.9 seconds, respectively. The design of Deep RC guarantees the smooth integration of scalable data frameworks, such as Cylon, with deep learning processes, exhibiting strong performance on cloud platforms and scientific HPC systems. By offering a flexible, high-performance solution for resource-intensive applications, this method closes the gap between data preprocessing, model training, and postprocessing. 

   more » « less
5. Deep RC: A Scalable Data Engineering and Deep Learning Pipeline

   Sarker, Arup; Alsaadi, Aymen; Halpern, Alexander; Tangella, Prabhath; Titov, Mikhail; Perera, Niranda; Staylor, Mills; Laszewski, Gregor von; Jha, Shantenu; Fox, Geoffrey (June 2025, Springer. JSSPP 2025: Job Scheduling Strategies for Parallel Processing)

   Significant obstacles exist in scientific domains including genetics, climate modeling, and astronomy due to the management, preprocess, and training on complicated data for deep learning. Even while several large-scale solutions offer distributed execution environments, open-source alternatives that integrate scalable runtime tools, deep learning and data frameworks on high-performance computing platforms remain crucial for accessibility and flexibility. In this paper, we introduce Deep Radical-Cylon(RC), a heterogeneous runtime system that combines data engineering, deep learning frameworks, and workflow engines across several HPC environments, including cloud and supercomputing infrastructures. Deep RC supports heterogeneous systems with accelerators, allows the usage of communication libraries like \texttt{MPI}, \texttt{GLOO} and \texttt{NCCL} across multi-node setups, and facilitates parallel and distributed deep learning pipelines by utilizing Radical Pilot as a task execution framework. By attaining an end-to-end pipeline including preprocessing, model training, and postprocessing with 11 neural forecasting models (PyTorch) and hydrology models (TensorFlow) under identical resource conditions, the system reduces 3.28 and 75.9 seconds, respectively. The design of Deep RC guarantees the smooth integration of scalable data frameworks, such as Cylon, with deep learning processes, exhibiting strong performance on cloud platforms and scientific HPC systems. By offering a flexible, high-performance solution for resource-intensive applications, this method closes the gap between data preprocessing, model training, and postprocessing. 

   more » « less