This job has closed.

Berkeley Lab · 4 weeks ago

Senior HPC Cluster Systems Administrator

Berkeley, CA

Full-time

Hybrid

Senior Level, Lead/Staff

$179K/yr - $218K/yr

12+ years exp

Berkeley Lab’s Information Technology Division is seeking a Senior HPC Cluster Systems Administrator to support the research community by maintaining Linux-based resources and high-performance computing cluster systems. The role involves extensive expertise in High Performance Computing infrastructure to facilitate groundbreaking research globally.

Research

No H1B

Responsibilities

Perform Linux system and HPC cluster maintenance and installations, operating system upgrades, system security hardening and intrusion detection, storage and file system management, system hardware, customization of user group working environment, troubleshooting, network monitoring, and crash recovery

Design, deploy, and manage scalable applications using Kubernetes, ensuring the availability, performance, and readiness of the Kubernetes infrastructure

Automate deployment, scaling, and management of containerized applications, and collaborating with DevOps and development teams to streamline CI/CD pipelines

Design, deploy, and manage the global storage platform to ensure high performance, massive scalability, reliability, and future-proof solutions

Support storage technologies such as Lustre, VAST, and networks

Resolve I/O issues related to business applications, including diagnosing and resolving complex storage, Linux, and networking challenges in a fast-paced environment

Research new storage management technologies, techniques, and provide recommendations

Participate in developing system administration, security, and network policies, documentation, and tools oriented towards efficient systems management

Participate in cluster support to staff and researchers, including initial installation, integration, and ongoing maintenance of Linux High-Performance Computing cluster systems. This includes travel to remote sites if as needed

Co-leading technical efforts with other senior system administrators in areas of HPC technologies such as job schedulers, high-performance interconnects, parallel file systems, cybersecurity, cluster management, container orchestration, VM infrastructure, networking, performance tuning, or data center planning

Co-leading group projects of small to medium size and complexity, to implement and deploy new computing technologies and associated services to the research community

Qualification

Linux system administrationHigh Performance ComputingKubernetes managementStorage system designPythonBashCI/CD toolsRed Hat derivativesInterpersonal skillsCommunication skillsProject management

Required

A Bachelor's Degree (or equivalent knowledge/training) in Computer Science, Engineering, or a related discipline, and a minimum of 12 years of relevant experience in Linux system administration within a large distributed computing environment, including experience providing systems and end-user support for multiple scientific or computational research groups or an equivalent combination of education and experience

Demonstrated ability to manage large-scale, performance-critical environments, including capacity planning, scaling, and optimization

Significant experience deploying, scaling, and managing Kubernetes clusters, with a strong understanding of its architecture (pods, deployments, services, ingress) and container orchestration. Proven proficiency with CI/CD tools like Jenkins or GitLab CI

Proven experience with Red Hat derivatives (CentOS, Scientific Linux, Rocky Linux), Debian, Ubuntu, and large-scale system and configuration management tools (Kickstart, Ansible, Puppet, Chef, Warewulf). Expertise in supporting standard services (NFS, LDAP, SMB, MySQL, Apache/Nginx HTTPD)

Strong HPC expertise, including Linux, job schedulers, high-performance interconnects, parallel file systems, cybersecurity, container orchestration, cluster management, VM infrastructure, networking, performance tuning, scientific application support, and data center planning

Proficiency in Python and Bash for building, optimizing, and debugging scientific codes (C, C++, Fortran, Java), including experience with compilers (GCC, Intel), debuggers, Makefiles, and version-control (git, Subversion)

Expertise in storage system design and optimization (Lustre, S3, VAST, Weka, Ceph, DDN), including a deep understanding of the storage stack (kernel to user space, including file systems, block storage, I/O schedulers, VFS), storage benchmarking, and performance tuning (throughput, latency, IOPS, workload-specific optimizations)

Excellent oral and written communication skills including experience organizing and presenting customer focused technical data, reports, and projects to audiences with varying degrees of technical expertise

Strong interpersonal skills including experience with research facilitation and project management in a multidisciplinary team environment