Job Description – System Administrator (Linux, Parallel Filesystems)
Company : Paroscale Technologies Pvt Ltd
Position Type : Full-time
About the Role
We are looking for highly skilled System Administrator Engineers with deep expertise in Linux systems and hands-on experience deploying, configuring, and troubleshooting Parallel File Systems (PFS) such as Lustre and / or BeeGFS .
The ideal candidate will have strong knowledge of distributed system architectures, high-performance network configurations, and performance tuning for large-scale storage environments.
Key Responsibilities
- Install, configure, and maintain Lustre , BeeGFS , or other parallel filesystem components (MDT, MDS, OSS / OST, clients).
- Manage Linux-based servers (Ubuntu, RHEL / CentOS, Rocky, etc.) used in HPC and distributed storage infrastructures.
- Configure and optimize network components such as InfiniBand , Ethernet , bonding , routing , VLANs , and TCP / IP tuning .
- Perform end-to-end troubleshooting of performance bottlenecks across compute, network, and storage layers.
- Implement and manage monitoring , logging , and alerting for production clusters.
- Work closely with engineering teams to support product development, testing, and deployment workflows.
- Execute system upgrades, kernel patching, and capacity planning.
- Automate repetitive tasks with Bash, Python, or Ansible.
- Perform benchmarking and performance analysis using tools like perf , IOR , MDTest , fio , etc.
- Document architectures, configurations, and operational procedures.
Required Skills & Experience
3–8 years of experience as a Linux System Administrator or HPC / Storage Engineer.Deep understanding of parallel filesystems (Lustre, BeeGFS preferred).Strong knowledge of Linux internals , kernel modules, and system performance tuning.Experience with distributed systems and high-speed networking (InfiniBand, RoCE, DPDK is a plus).Hands-on experience with benchmarking and profiling tools (perf, IOR, MDTest, fio).Experience with virtualization or container environments (KVM, Docker, Kubernetes is a plus).Strong troubleshooting skills, ability to analyze logs, core dumps, and system metrics.Excellent communication and documentation skills.Why Join Us?
Work with deep-tech engineers solving challenging problems in distributed storage.Opportunity to contribute to high-performance computing, parallel filesystems, and large-scale infrastructure.Collaborative and growth-oriented environment with exposure to open-source projects.