HPC Network Engineer

HPC Network Engineer (Private Cloud) Whitehall Resources require an HPC Network Engineer (Private Cloud) to work with a key client on a 6 month initial contract.

• Inside IR35.
• 2 days per week on site in Cambridge. HPC Network Engineer (Private Cloud) Job Overview: This is a fantastic opportunity to join a newly formed Private Cloud team. We are responsible for the design, deployment and operations of a new greenfield on-prem private cloud based on OpenStack. The Private Cloud HPC Network Engineer will help design, build and evolve the high-performance network architecture underpinning our Private Cloud platform that will be used for HPC workloads. The role combines deep data-centre and HPC networking expertise with modern cloud networking, software-defined infrastructure and automation. As part of the Private Cloud team, you will collaborate closely with compute, storage, platform engineering teams to deliver a scalable, resilient and highly automated cloud infrastructure. Responsibilities:
• Design and evolve network architecture, standards and patterns for large-scale Private Cloud Openstack and HPC environments. Develop scalable Leaf-Spine architectures using technologies including BGP, EVPN, VXLAN and ECMP.
• Design networking for high-performance compute, storage and latency-sensitive engineering workloads. Evaluate technologies including SR-IOV, SmartNICs, DPUs and high-performance Ethernet.
• Integrate physical network infrastructure with OpenStack networking services including Neutron, OVN and ML2. Define network architecture for virtual machines, bare-metal workloads and Kubernetes platforms.
• Perform performance analysis, network optimisation and lead technical evaluation and benchmarking of networking technologies and vendors.
• Lead network security architecture and threat modelling while also leading complex cross-domain issues spanning compute, storage, networking and HPC applications. Required Skills and Experience:
• Strong experience designing and supporting high-performance or large-scale compute environments using RoCE, RDMA or Infiniband in a Cisco NX-OS and/or Arista EOS .
• Deep understanding of BGP routing and network architectures, Leaf-Spine network, Layer 2 and Layer 3, ECMP, EVPN and VXLAN and network resilience. With exposure to newer technologies such as SmartNICs, DPUs & NVMe-over-Fabrics..
• Practical experience of OpenStack network services and developing integrated network automation and API driven interfaces.
• Strong HPC network observability, troubleshooting and performance tuning skills. “Nice To Have” Skills and Experience:
• Strong Linux knowledge and Kubernetes networking.
• Exposure to SR-IOV and DPDK.
• Demonstrable hands-on Python, Ansible and Terraform experience.
• Using SRE principles in the creation of SLI/SLO and error budgets