Greater Chicago Area
Ph.D. in computational physics. 14 years of experience in HPC/Linux. 9 years of experience in high performance and grid computing administration. Strong passion for HPC system engineering.
Develop and administrate the FIU HPC environment consisting of 100+ servers, 30 Nvidia GPU accelerators, FDR infiniband network, DDN GPFS storage, LSF scheduler, visualization systems, and 200 applications. Support 50+ active users. Serve as OSG FIU site gridadmin. Success in TechFee funding proposal. –– Deploy and manage via xCAT the campus cluster. Diskful/diskless images, access security, local Yum repository, postscript automation, Nagios monitoring. –– Install/compile applications including Matlab, R, Namd, Molpro, NWchem, ANSYS, Abaqus, and Blast. Module software environment management. Help users for submission work flow and job troubleshooting. –– Administrate the resource manager IBM Platform LSF. Implement queue, host group, user group to maximize resource utilization and assure QoS agreements with investors. –– Administrate the 240TB DDN SFA Gridscaler GPFS storage. filesystem, fileset, snapshot, replication, quota, policy, IB verb, I/O diagnosis and benchmark. SFA physical disk, virtual disk, interconnect, and embedded KVM images. –– Develop Mellanox FDR Infiniband network. IB switch, subnet manager, OFED installation, basic performance benchmark. –– Develop the remote visualization environment based on NICE software. EnginFrame webportal to start VNC remote desktop; LSF integration to loadbalance VNC sessions. Wrote Python and bash codes for convenient features such as session timer and session closing alert. –– Design and build an ultra-high resolution 140-inch video wall, totaling 18 million pixels by leveraging Nvidia Mosaic technology to seamlessly span application, image, and video across nine high-resolution panels in order to help scientists to “see” the large-scale data. –– Serve as the gridadmin to connect FIU HPC cluster as a resource provider to Open Science Grid. X.509 certificate, MyProxy, GridFTP, Condor-CE, Globus. –– Write University TechFee funding proposal approved by CIO for cluster expansion.
Participate the NSF-founded Extending Science Through Enhanced National Cyberinfrastructure (ExTENCI) project.
Investigate quantum phase transitions in magnetic impurity models