DEVOPS ENGINEER M/F DISTRIBUTED STO-RAGE/PROCESSING ARCHITECTURE FOR SKA DATA H/F

CEA; Gif-sur-Yvette; Sur place; EUR 70 000 - 100 000; Plein temps

Résumé du poste

CEA Saclay is seeking a DevOps Engineer to deploy a distributed storage and processing infrastructure for SKA data within the NumPEx programme, starting at Saclay. You will deploy the SRCNet stack across heterogeneous hardware, scale a demonstrator at the Paris‑Saclay mesocentre, and monitor operational conditions of services.

You will collaborate with engineers and physicists, applying Kubernetes, Terraform/Ansible, and OpenStack, while ensuring security and performance in a complex HPC

Qualifications

• Master's or engineering degree in software or related field.
• Deep Kubernetes experience with ArgoCD, Gateway API, and Cilium.
• Strong DevOps and CI/CD practices using GitLab.
• Containerisation with security best practices and Linux proficiency.
• Experience with infrastructure‑as‑code (Terraform/Ansible) and OpenStack.
• Monitoring with Prometheus/Grafana and scripting (bash/python).

Responsabilités

• Deploy SRCNet software stack across distributed hardware.
• Configure and maintain FR SKA Regional Centre infrastructure.
• Deploy and scale a demonstrator at Paris‑Saclay mesocentre.
• Monitor and maintain operational conditions of deployed services.
• Propose improvements to existing processes.

Connaissances

Kubernetes

GitLab CI/CD

Monitoring

Linux

Scripting (Python, Bash)

Formation

Master's or engineering degree

Outils

ArgoCD

Gateway API

Cilium

RKE2

OpenStack

Terraform

Ansible

Prometheus

Grafana

Helm

Rucio

XRootD

Indigo-IAM

Description du poste

DEVOPS ENGINEER M/F DISTRIBUTED STORAGE/PROCESSING ARCHITECTURE FOR SKA DATA H/F

Contract: CDD

Locations: Saclay, Gif-sur‑Yvette

Institution: IRFU, DEDIP, CEA Saclay.

OVERVIEW

Irfu and DEDIP are leading the PEPR (Priority Research Programme and Equipment) NumPEx (High‑Performance Computing for Exascale) programme, funded by the French government’s PIA and headed by the CEA, INRIA and the CNRS. The scientific objectives are to develop innovative methods and software tools for future exascale machines and to support application domains that use HPC resources. In this context, DEDIP is recruiting a DevOps engineer to deploy a distributed storage and processing infrastructure for data from the Square Kilometer Array (SKA) radio telescope, currently under construction in Australia and South Africa. This national infrastructure (French SKA Regional Centre) will be part of an international network (SRC Network) of data centres. The network will manage an annual data volume of around 700 PB, relying on proven and innovative high‑performance technologies.

You will join a multidisciplinary team of computer engineers and physicists working within the PEPR NumPEx programme, in partnership with several processing mesocentres, national computing centres and laboratories. Your main tasks will be to deploy the software stack provided by SRCNet, make technological choices suited to the distributed heterogeneous hardware, and continue to monitor and maintain the operational conditions of the services deployed by the FRSRC. The first step is to set up a functional demonstrator at the Paris‑Saclay mesocentre (in collaboration with IJCLab), before scaling the architecture to other data centres.

In addition, you will be a driving force in proposing improvements to existing processes.

RESPONSIBILITIES

• Deploy the SRCNet software stack across distributed heterogeneous hardware.
• Configure and maintain the French SKA Regional Centre (FRSRC) infrastructure.
• Deploy and scale a functional demonstrator at the Paris‑Saclay mesocentre.
• Monitor and maintain operational conditions of deployed services.
• Propose improvements to existing processes.

QUALIFICATIONS

• Master’s or engineering degree in software engineering.
• In‑depth knowledge of Kubernetes, including ArgoCD, Gateway API, cilium, and RKE2.
• Expertise in DevOps, configuration management with Git, and CI/CD with GitLab.
• Solid experience in application containerisation following best security practices.
• Experience with monitoring tools for Kubernetes (Prometheus, Grafana).
• Proficiency in Linux.
• In‑depth understanding of functional and integration testing.
• Experience with infrastructure‑as‑code tools (Terraform, Ansible).
• Experience deploying in cloud environments (OpenStack).
• Knowledge of Kubernetes application packaging (Helm).
• Knowledge of network issues, monitoring, security constraints, and identity & authorisation management (Indigo‑IAM).
• Experience with Rucio storage virtualization and the XRootD protocol.
• Proficiency in one or two scripting languages (bash, Python).

SOFT SKILLS

• Enjoys working in multidisciplinary teams and is proactive.
• Pragmatic, takes initiative, and can calmly tackle complex problems under constraints.
• Curious, adaptable, autonomous, with good interpersonal skills.
• Excellent written and oral communication in French and English; proficient in technical English.

START DATE

01/09/2026