Postdoctoral Research Associate in HPC File Systems and Storage
- Oak Ridge, TN
ORNL’s computational expertise is built on a foundation of computer science, mathematics, and “big data”—or data science. The projects we undertake run the gamut from basic to applied research, and our ability to efficiently apply the massive computing power available at ORNL across a range of scientific disciplines sets us apart from other computing centers. We have decades of experience in developing applications to support basic science research in areas ranging from chemistry and materials science to fission and fusion, and we apply that expertise to solving problems in a number of other areas.
Oak Ridge National Laboratory is home to Titan, a Cray XK7 high-performance computing (HPC) system which debuted atop the list of world’s fastest supercomputers in November 2012 with a calculating speed of 17.59 petaflops (quadrillion floating point operations per second). This blazingly fast supercomputer will decrease time to solution, increase complexity of models, and improve realism of simulations of subjects ranging from development of advanced materials to cleaner combustion of fuels.
ORNL is also home to Atlas, a Lustre file system used for computational resources. As an extremely high-performance system, Atlas has over 26,000 clients, providing 32 petabytes of disk space and can move data at more than 1 Terabyte per second.
HPSS is the archival mass-storage resource at ORNL and consists of robotic tape and disk storage components, Linux servers, and associated software. Incoming data is written to disk and later migrated to tape for long term archival. As storage, network, and computing technologies continue to change, ORNL’s storage system evolves to take advantage of new equipment that is both more capable and more cost-effective.
The Extreme Scale Systems Center in conjunction with the Computer Science Research Group in the Computer Science and Mathematics Division, at the Oak Ridge National Laboratory has an opening for a Postdoctoral Researcher in the field of high performance file systems and storage.
The job will involve collaborating with vendors, national laboratories, and universities in leveraging experimental hardware and software configurations. Additionally, the job will involve improving the performance and adding new capabilities to existing software packages as well as developing original software packages and benchmarks.
Additional job responsibilities include:
- Research, design, and implement techniques for building a fault-tolerant, wide area runtime system.
- Research, design, and implement techniques for efficiently and reliably moving large data sets over long fat networks (LFN).
- Research, design, and implement software techniques for efficiently creating and interacting with file system metadata.
- Publish in peer-reviewed journals and conference proceedings
Minimum Qualifications Required
Ph.D. in Computer Science, Computer Engineer, Mathematics or related field
Strong programming skills
Programming in C and/or C++
Multithreading using POSIX threads
Additional desirable skills include experience or familiarity with:
• Experience with Python
• Network programming using TCP sockets
• Linux kernel/driver development
• HPC/cluster network fabrics and programming APIs
• Parallel and distributed file systems
• Big Data programming models and systems such as Hadoop
• Performance analysis, measurement, and/or modeling of distributed systems
This position requires access to technology that is subject to export control requirements. Successful candidates must be qualified for such access without an export control license. As a result, U.S. Citizenship or LPR (Lawful Permanent Residence) is required.
Applicants cannot have received the most recent degree more than five years prior to the date of application and must complete all degree requirements before starting their appointment. This position is a temporary, full-time assignment not to exceed 24 months. Certain exceptions may be considered.