The landscape of gpu-centric communication
D Unat, I Turimbetov, MKT Issa, D Sağbili… - arXiv preprint arXiv …, 2024 - arxiv.org
In recent years, GPUs have become the preferred accelerators for HPC and ML applications
due to their parallelism and fast memory bandwidth. While GPUs boost computation, inter …
due to their parallelism and fast memory bandwidth. While GPUs boost computation, inter …
Boosting Data Center Performance via Intelligently Managed Multi-backend Disaggregated Memory
J Wang, H Yang, C Li, Y Zhuansun… - … Conference for High …, 2024 - ieeexplore.ieee.org
Existing disaggregated memory (DM) systems face a problem of underutilized far memory
bandwidth, which greatly limits the data throughput when processing data-intensive …
bandwidth, which greatly limits the data throughput when processing data-intensive …
Bandwidth-Effective DRAM Cache for GPU s with Storage-Class Memory
We propose overcoming the memory capacity limitation of GPUs with high-capacity Storage-
Class Memory (SCM) and DRAM cache. By significantly increasing the memory capacity …
Class Memory (SCM) and DRAM cache. By significantly increasing the memory capacity …
GRIT: Enhancing Multi-GPU Performance with Fine-Grained Dynamic Page Placement
Multi-GPU systems have become popular to cater to the growing demands for high
parallelism and large memory capacity. However, the delivered performance is constrained …
parallelism and large memory capacity. However, the delivered performance is constrained …
Excavating the potential of graph workload on rdma-based far memory architecture
Disaggregated architecture brings new opportunities to memory-consuming applications like
graph processing. It allows one to outspread memory access pressure from local to far …
graph processing. It allows one to outspread memory access pressure from local to far …
Fargraph+: Excavating the parallelism of graph processing workload on RDMA-based far memory system
Disaggregated architecture brings new opportunities to memory-consuming applications like
graph processing. It allows one to outspread memory access pressure from local to far …
graph processing. It allows one to outspread memory access pressure from local to far …
Fine-grain Quantitative Analysis of Demand Paging in Unified Virtual Memory
The abstraction of a shared memory space over separate CPU and GPU memory domains
has eased the burden of portability for many HPC codebases. However, users pay for ease …
has eased the burden of portability for many HPC codebases. However, users pay for ease …
GPUVM: GPU-driven Unified Virtual Memory
N Nazaraliyev, E Sadredini… - arXiv preprint arXiv …, 2024 - arxiv.org
Graphics Processing Units (GPUs) leverage massive parallelism and large memory
bandwidth to support high-performance computing applications, such as multimedia …
bandwidth to support high-performance computing applications, such as multimedia …
HyFarM: Task Orchestration on Hybrid Far Memory for High Performance Per Bit
Tapping into secondary memory resources, ie, far memory (FM), has shown huge potential
to improve the cost-efficiency of data centers. Recent advances in both storage-based …
to improve the cost-efficiency of data centers. Recent advances in both storage-based …
Early-Adaptor: An Adaptive Framework forProactive UVM Memory Management
Unified Virtual Memory (UVM) relieves programmers of the burden of memory management
between CPU and GPUs. However, the use of UVM can lead to performance degradation …
between CPU and GPUs. However, the use of UVM can lead to performance degradation …