26 to 29 of 29 InfiniBand Jobs in London

authority for the company. You own the fabric from architecture through day-2 operations. Responsibilities Design and operate lossless, RDMA-capable fabrics (e.g. RoCEv2, InfiniBand) for GPU compute and storage traffic, including QoS, congestion control and buffer tuning at scale Build and manage leaf-spine data centre fabrics, with routed … Linux networking stack, not just a vendor CLI Practical RDMA fabric experience: lossless Ethernet (e.g. RoCEv2 with PFC/ECN/DCQCN tuning) or InfiniBand, understanding why lossless behaviour matters for GPU workloads Network automation as a working practice: scripting (e.g. Python), configuration management (e.g. Ansible), config generation from ...

decade - this is the place. THE ROLE: HPC ENGINEER We're deploying more and more large-scale NVIDIA GPU clusters connected by InfiniBand, and the gap between a cluster that's built and one that delivers its expected training performance is where customer experience is won or lost. … with NVIDIA and OEMs on hardware faults and RMAs Run production HPC clusters day to day, including Slurm, GPU drivers and the CUDA stack, InfiniBand fabrics (subnet management, monitoring, firmware) and the parallel storage clusters depend on Tune clusters for AI training and HPC workloads, from NCCL settings and GPUDirect ...

looking for with this role: You will: Design and develop scalable AI network and fabric architectures based on customer and data centre requirements, leveraging InfiniBand, Ethernet, AFL, and emerging networking technologies. Lead the end-to-end architecture of AI infrastructure, including GPU-to-GPU connectivity, server-to-storage communication … applications while advancing innovation in high-density, liquid-cooled data centres. Essential Requirements Deep expertise in AI and data center networking technologies, including Ethernet, InfiniBand, AI fabrics, switches, NICs, DPUs, optical networking, and large-scale fabric design. Strong understanding of network topologies and architectures optimized for AI training and inference ...

centre, power and ongoing operational expenditure Evaluate and recommend data centre/colocation partners capable of supporting high-density GPU racks Define networking topology (InfiniBand/Ethernet fabric) to optimise for distributed training and inference workloads Liaise with hardware vendors, OEMs and data centre providers to validate specifications and negotiate … NVIDIA H100/H200/B200/B300 class systems), server/rack design, and data centre infrastructure Methodology/Protocols: Strong grasp of InfiniBand/RDMA networking, NVLink/NVSwitch topologies, and cluster interconnect design for distributed compute workloads Soft Skills: Confident operating at boardroom level, able to translate ...