2022
Aquila: A unified, low-latency fabric for datacenter networks
19th USENIX Symposium on Networked Systems Design and Implementation (NSDI ’22) · Renton, WA, April 2022
Research
My doctoral research at Purdue, advised by Charles Killian, was on how to build distributed systems that are both parallel and reliable without making them harder to write. The dissertation introduced a unified model for multi-threaded, fault-tolerant distributed systems frameworks; the pieces below — InContext, MaceKen and EventWave — are its instances, all built on the Mace language and runtime.
At Google my work has been in software-defined networking for data centers: the Orion control plane, Aquila’s low-latency fabric, and the routing engine that programs Google’s switches.
Publications
2013
EventWave: Programming Model and Runtime Support for Tightly-Coupled Elastic Cloud Applications
ACM Symposium on Cloud Computing (SoCC ’13) · Santa Clara, CA, October 2013
2012
Composable Reliability for Asynchronous Systems
USENIX Annual Technical Conference (USENIX ATC ’12) · Boston, MA, June 2012
2011
InContext: Simple Parallelism for Distributed Applications
20th ACM Symposium on High-Performance Parallel and Distributed Computing (HPDC ’11) · San Jose, CA, June 2011
2012
The combined manpower planning and preventive maintenance strategies by aggregate planning for the service division of vending machines
Journal of the Operational Research Society · Vol. 63, No. 8, pp. 1034–1050
2010
A Unified Framework for Trees, Multi-Dimensional Scaling and Planar Graphs
arXiv · arXiv:1012.5887
2009
Sensitivity Analysis on the Operational Strategies for Maintenance (민감도 분석을 통한 정비부서의 운영 전략 연구)
Korean Management Review (경영학연구) · Vol. 38, No. 2, pp. 555–582