Cloud HPC Network Engineer

Allscreens Nationwide Ltd; Cambridge; On-site; GBP 70,000 - 110,000; Full time

Job summary

Allscreens Nationwide Ltd in Cambridge is seeking an experienced Private Cloud HPC Network Engineer to join a newly formed team responsible for designing, deploying and operating a greenfield on-premises OpenStack private cloud environment.

You will design and evolve Leaf-Spine networks for HPC workloads, implement BGP, EVPN, VXLAN and ECMP, and integrate Neutron/OVN/ML2 with compute, storage and Kubernetes environments to deliver a scalable, automated cloud platform.

Qualifications

• Experience designing HPC or large-scale compute networks.
• Proficient with Leaf-Spine fabrics and data center networking.
• Strong knowledge of BGP/EVPN/VXLAN/ECMP and related protocols.

Responsibilities

• Design and evolve network architecture for large OpenStack private cloud and HPC environments.
• Develop scalable Leaf-Spine networks using BGP, EVPN, VXLAN and ECMP.
• Integrate physical networking with OpenStack Neutron/OVN/ML2 and compute/storage teams.
• Support network performance tuning and observability across HPC environments.
• Collaborate with compute, storage and platform engineering to resolve cross-domain issues.

Skills

HPC networking

RoCE/RDMA/InfiniBand

Leaf-Spine architectures

Troubleshooting & observability

Tools

OpenStack Neutron

OpenStack OVN

ML2

Job description

We are seeking an experienced Private Cloud HPC Network Engineer to join a newly formed team responsible for the design, deployment and operation of a greenfield on-premises OpenStack private cloud environment. You will play a key role in building and evolving high-performance network infrastructure supporting HPC workloads, working across compute, storage and platform engineering teams to deliver a scalable, resilient and highly automated cloud platform.

• Start Date: ASAP

Key Responsibilities:

• Design and evolve network architecture for large-scale OpenStack private cloud and HPC environments
• Develop and support scalable Leaf-Spine network architectures using BGP, EVPN, VXLAN and ECMP
• Design networking solutions for high-performance compute, storage and latency-sensitive workloads
• Integrate physical networking infrastructure with OpenStack technologies including Neutron, OVN and ML2
• Define network architectures supporting virtual machines, bare-metal platforms and Kubernetes environments
• Perform network optimisation, benchmarking, observability and performance tuning across HPC environments
• Work closely with compute, storage and platform engineering teams to resolve complex cross-domain infrastructure challenges

Key Requirements:

• Strong experience designing and supporting HPC or large-scale compute environments using RoCE, RDMA and/or InfiniBand
• Strong knowledge of Cisco NX-OS and/or Arista EOS networking platforms
• Deep understanding of BGP, EVPN, VXLAN, ECMP, Layer 2/Layer 3 networking and Leaf-Spine architectures
• Experience with OpenStack networking services including Neutron, OVN and ML2
• Strong troubleshooting, network observability and performance tuning experience within HPC or data centre environments
• Experience with network automation, API-driven infrastructure and software-defined networking

Additional Information:

• Exposure to Kubernetes networking, Linux administration, Python, Ansible and Terraform would be highly beneficial
• Experience with SR-IOV, DPDK, SmartNICs, DPUs and NVMe-over-Fabrics is desirable
• Knowledge of network security architecture, threat modelling and SRE principles would be advantageous

Interview Process:

Two-stage interview process