20 results for “GPU sharing”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
Zhihao Xu, Hao Zhong, Zeting Zhou, Yuhang Xu +8 more
This paper proposes Gleam, a framework for efficient GPU sharing across local-area CUDA devices, reducing bandwidth overhead, improving API call latency, and ensuring context consistency.
Yuncheng Yao, Zhenzhou Qi, Junyao Zheng, Chung-Hsuan Tung +2 more
This paper presents Beaver, a GPU sharing system that jointly manages compute and memory resources to protect the vRAN's strict processing deadline while serving AI workloads.
This paper presents Festina, a profiling-guided, power-aware control plane for minimizing energy consumption in serverless large language model (LLM) serving.
This paper presents ScaleRAN, a runtime for elastic GPU resource management in AI-RAN, enabling dynamic partitioning of GPU resources between RAN and AI workloads.
Shushu Yi, Wenbo Wu, Guoci Chen, Junrong Zhu +5 more
GNStor is a GPU-native AFA system that enables direct GPU access to remote AFA without CPU intervention.
Michael Wasef, Gongjin Sun, Edmund Au, Ranjitha Gopalakrishna +14 more
CXL Memory Module - Memory Pooling (CMM-MP) is a novel memory-pooling architecture for emerging memory-intensive workloads.
This paper analyzes data access patterns in large language model workloads on multi-partition GPUs and presents optimization strategies based on inter-workgroup sharing patterns.
Melih Elibol, Jared Roesch, Isaac Gelado, Eric Buehler +1 more
cuTile Rust provides a system for safe, idiomatic GPU kernel authoring in Rust.
This paper introduces Eidola, a scalable extension to the gem5 simulation framework for modeling inter-GPU communication traffic.
Shuheng Gao, Fan Guo, Yongkun Li, Ziye Yue +2 more
gKSM is a transparent GPU memory deduplication framework for serverless LLM inference that reduces GPU memory usage and improves throughput.
This paper presents AgileOS, a GPU operating-system layer for protected CUDA services.
This paper presents a rustc-based GPU compilation framework for Rust, ensuring memory safety and efficiency in high-performance GPU programming.
Tangram enables the use of existing LLM parallelizers in heterogeneous GPU clusters by decoupling parallelization planning from GPU heterogeneity.
Xinwei Qiang, Yifan Hu, Shixuan Sun, Jing Yang +5 more
GF-DiT: A policy-programmable runtime for elastic DiT serving
This paper identifies suboptimalities in default GPU-first scheduling for AI tool workloads and proposes an agentic scheduler that adaptively assigns tools to GPU or CPU based on runtime factors.