...
- Oak Ridge National Laboratory. Remote Contract
- MPI and libfabric Design and Development
● Perform project management tasks
● Write project requirements and high level designs
● Integrate and customise Open MPI and libfabric for next generation Super Computers
● Work with the open source community to upstream relevant features - HPC Systems Engineer
- MPI and libfabric Design and Development
March 2018– October 2021
- Data Direct Networks (DDN). Remote
- Lustre Networking Technical Lead, Open Source Development
- Dynamic LNet Configuration (DLC) project
- Designed and implemented Multi-Rail for LNet (https://wiki.whamcloud.com/x/tZmCBw)
- Multi-Rail Health Monitoring (https://wiki.whamcloud.com/x/HR1eC)
- Multi-Rail Lustre Routing (https://wiki.whamcloud.com/x/IB1eC)
- User Defined Selection Policy (https://wiki.whamcloud.com/x/t5mCBw)
- Distributed Test Infrastructure (https://wiki.whamcloud.com/x/ioH5Bw)
- Worked along side NASA Ames, ORNL, LLNL, SNL, Stanford University, Harvard, Brigham Young University, as well as many other public and private entities to gather Luster Networking Requirements, make recommendations and resolve issues they run into with their Super Computing Clusters
- Presented at the Open Fabrics and Luster User Groups conferences multiple times (https://youtu.be/07EmqaeD63E?list=PLs1xv9ddvod4sCVakpKpdD9Cr28vcd5B8)
- Worked on creating a Lustre front end to DDN's RED, a Key/Value Storage system.
- Lustre Networking Technical Lead, Open Source Development
...