5 papers
Double-P: Hierarchical Top-P Sparse Attention for Long-Context LLMs
Wentao Ni, Kangqi Zhang, Zhongming Yu +7
As long-context inference becomes central to large language models (LLMs), attention over growing key-value caches emerges as a dominant decoding bottleneck, motivating sparse atte…
TraCT: Disaggregated LLM Serving with CXL Shared Memory KV Cache at Rack-Scale
Dongha Yoon, Younghoon Min, Hoshik Kim +2
Disaggregated LLM serving improves resource efficiency by separating the compute-intensive prefill phase from the latency-critical decode phase. However, this architecture introduc…
Accelerating Sparse Matrix-Matrix Multiplication on GPUs with Processing Near HBMs
Shiju Li, Younghoon Min, Hane Yie +5
Sparse General Matrix-Matrix Multiplication (SpGEMM) is a fundamental operation in numerous scientific computing and data analytics applications, often bottlenecked by irregular me…
Cocoon: A System Architecture for Differentially Private Training with Correlated Noises
Donghwan Kim, Xin Gu, Jinho Baek +6
Machine learning (ML) models memorize and leak training data, causing serious privacy issues to data owners. Training algorithms with differential privacy (DP), such as DP-SGD, hav…
cMPI: Using CXL Memory Sharing for MPI One-Sided and Two-Sided Inter-Node Communications
Xi Wang, Bin Ma, Jongryool Kim +3
Message Passing Interface (MPI) is a foundational programming model for high-performance computing. MPI libraries traditionally employ network interconnects (e.g., Ethernet and Inf…