Geared toward experienced Lustre developers comfortable with kernel-level and distributed systems performance optimization, located in Santa Clara with a focus on upstream contributions and architectural modernization.
Derived from job-description analysis by Serendipath's career intelligence engine.
Original posting from DDN via Ashby
WHAT YOU’LL DO
- Contribute to Lustre architecture and lead implementation of new features across core subsystems including MDS/OSS, LDLM, DNE, DoM, FLD, layout, snapshot, quota, security, and recovery
- Write high-performance, production-grade code across the Lustre kernel module, user-space utilities, and supporting libraries
- Own the full development lifecycle for major features, from design and coding through testing, performance tuning, and upstream patch submission
- Improve scalability for multi-petabyte Lustre environments, including metadata performance, I/O path optimization, and integration with modern interconnects and storage fabrics
- Modernize Lustre internals through kernel API updates, locking improvements, RDMA enhancements, and cloud/container-oriented capabilities
- Prototype and evaluate emerging technologies such as NVMe-oF, CXL, and other next-generation storage approaches for potential Lustre integration
- Strengthen regression, stress, and performance validation using tools such as IOR, mdtest, fio, and custom Lustre test tooling
- Perform deep code-level performance analysis using Lustre tracing tools, perf, lockstat, and kernel profilers
- Work closely with Principal Engineers, hardware teams, and application owners to translate real workload needs into practical Lustre improvements
- Mentor other developers and contribute design docs, technical guides, and engineering reviews
WHAT WE’RE LOOKING FOR
- 7+ years of progressive, hands-on Lustre software development experience
- A strong track record of meaningful contributions to the Lustre codebase in production or open-source environments
- Demonstrated success designing, implementing, and upstreaming Lustre features that made it into major releases
- Deep understanding of Lustre internals, including MDS/OSS architecture, LDLM locking and recovery, journaling, DNE, DoM, FLD, layout, quota, and security
- Expert-level C programming and strong Linux kernel development experience
- Strong knowledge of distributed systems, parallel file systems, and high-performance I/O
- Experience with Lustre build, development, and debugging tooling such as lctl, llstat, strace, perf, and kernel tracers
- Python and/or Bash skills for tooling and automation
- Bachelor’s or Master’s degree in Computer Science, Computer Engineering, or a related field
- Clear written and verbal communication skills, with the ability to explain complex technical concepts effectively
NICE TO HAVE
- 7+ years in HPC or storage software development
- Visible contributions to the Lustre open-source community, including accepted patches, LUG participation, or public technical talks
- Experience running or developing Lustre in cloud environments such as AWS FSx for Lustre, Azure Lustre, or custom cloud deployments
- Familiarity with adjacent systems such as Ceph, DAOS, BeeGFS, WekaIO, or GPFS
- Exposure to RDMA, NVMe-oF, CXL, computational storage, or AI/ML I/O workloads
- Conference presentations, technical publications, or other public engineering contributions
YOU’LL LIKELY BE A STRONG FIT IF YOU…
- Have built inside Lustre itself, not just around it
- Enjoy solving hard kernel, performance, and distributed systems problems
- Care about code quality, upstream contribution, and long-term architectural impact
- Want to work on infrastructure that operates at real scale under real performance constraints
THIS ROLE IS PROBABLY NOT THE RIGHT FIT IF YOU…
- Have mainly supported or administered Lustre rather than developed it
- Come from general storage or kernel backgrounds without clear Lustre code contribution history
- Prefer high-level application engineering over low-level systems and kernel work
Source: DDN careers (Ashby)