Sung-Han Lin

Sung-Han Lin is a performance and capacity engineer at Meta. He is interested in improving the performance of large-scale Deep Learning systems via investigating the bottleneck of data pipeline and optimizing system resource utilizations. He received his Ph.D. degree from the University of Southern California (USC) in 2017. His thesis topic is optimizing resource allocations in large-scale distributed systems, where one of the applications is improving the throughput of large-scale distributed training.