3 articles tagged with #rdma. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
Researchers have developed dmaplane, a Linux kernel module that provides buffer orchestration for AI workloads, addressing the gap between efficient data transport and proper buffer management. The system integrates RDMA, GPU memory management, and NUMA-aware allocation to optimize high-performance AI data paths at the kernel level.
OpenURMA is the first open-source implementation of Huawei's Unified Bus (UB) protocol, a 2025 specification designed to overcome RDMA bottlenecks at the network interface. The implementation demonstrates 4.37x lower latency and 2.80x higher throughput compared to RoCEv2, while consuming only 14% of FPGA resources, offering a potential architectural shift for datacenter networking.
Researchers have developed NCCL EP, a new communication library for Mixture-of-Experts (MoE) AI model architectures that improves GPU-initiated communication performance. The library provides unified APIs supporting both low-latency inference and high-throughput training modes, built entirely on NVIDIA's NCCL Device API.
