3 days ago
Remote, Germany +3 moreSenior
Responsibilities
- Engineer and integrate Lustre and NFS storage systems for large-scale HPC and AI clusters.
- Analyze performance and reliability across filesystem clients, Linux kernel paths, network protocols, servers, and storage devices.
- Develop, review, test, and deploy software fixes and performance improvements with internal teams, storage partners, and open-source maintainers.
- Characterize application I/O patterns including training-data access, checkpointing, restart, metadata-intensive operations, and concurrent workloads.
- Build reproducible benchmarks, diagnostic tools, and regression tests for efficiency, latency, metadata performance, scalability, and contention behavior.
- Lead storage qualification and commissioning, including validation of functionality, data integrity, failure recovery, and sustained cluster-scale performance.
- Define and implement storage placement, caching, and data lifecycle policies based on workload and operational evidence.
Requirements
- Deep experience with Lustre, NFS, or comparable shared-storage systems, including implementation-level debugging and Linux cluster integration.
- Strong C programming and Linux debugging skills, including experience modifying and diagnosing kernel or user-space storage software.
- Understanding of filesystem and distributed-system behavior, including metadata operations, caching, consistency, locking, concurrency, client/server communication, and recovery.
- Experience analyzing application access patterns and tuning filesystem or storage-stack behavior for performance.
- Ability to take complex performance or correctness issues through reproduction, root-cause analysis, implementation, and validation.
- Proficiency in Python and shell scripting for diagnostic, test automation, and analysis tools.
- Ability to independently lead technical efforts, influence cross-team design decisions, and communicate technical tradeoffs clearly.
- BS or MS in Computer Science, Engineering, Mathematics, Physics, or equivalent practical experience.
- 12+ years of relevant experience.
- Preferred experience contributing to Lustre, Linux filesystems, NFS implementations, or other storage projects.
- Preferred expertise in Lustre metadata and client I/O, Linux VFS and page cache, NFS internals, or distributed locking and recovery.
- Preferred experience with hierarchical storage management, policy engines, filesystem-to-object-storage migration, RDMA storage transports, GPUDirect Storage, GPU workload optimization, storage security, or multi-tenancy.
About Nvidia
Nvidia designs and sells GPUs and accelerated computing platforms for data centers, AI/ML, graphics, gaming, and automotive, monetizing through hardware, software platforms (CUDA, AI frameworks), and systems like DGX and networking. Customers include cloud providers, enterprises, researchers, and OEMs. Founded in 1993 and headquartered in Santa Clara, it is a public company traded on NASDAQ under NVDA.