Principles and Practice of Parallel Programming

• GLUMIN: Fast Connectivity Check Based on LUTs For Efficient Graph Pattern Mining
• Mario: Near Zero-cost Activation Checkpointing in Pipeline Parallelism
• PC Member in Program Committee within the Main Conference-track
• COMPSO: Optimizing Gradient Compression for Distributed Training with Second-Order Optimizers
• POSTER: FastBWA: Practical and Cost-Efficient Genome Sequence Alignment Pipeline
• POSTER: Magneto: Accelerating Parallel Structures in DNNs via Co-Optimization of Operators