Khiem Ngo

Khiem Ngo

Research Scientist

Meta Platforms, Inc.

Biography

I received my PhD in Computer Science from Princeton University, where I was advised by Professor Wyatt Lloyd in the S* Network Systems (SNS) group. Prior to Princeton, I was a member of the Netwoked Systems Laboratory at USC. I previously received my BComp in Computer Engineering from the School of Computing at NUS. I also worked at NUS and ADSC for a few years before starting my PhD.

I am generally interested in the design, implementation and evaluation of large-scale distributed systems. During my PhD, I developed two systems: Copilot replication, a 1-slowdown-tolerant consensus protocol, and K2, a causally-consistent, partially-replicated storage system.

Interests
  • Fault Tolerance
  • Consensus
  • Storage Systems
  • Consistency
Education

  • PhD in Computer Science, 2021

    Princeton University

  • MA in Computer Science, 2019

    Princeton University

  • BComp in Computer Engineering, 2010

    National University of Singapore

Experience

 
 
 
 
 
Meta Platforms, Inc.
Research Scientist
Meta Platforms, Inc.
Sep 2021 – Present Menlo Park, CA
 
 
 
 
 
Princeton University
Graduate Research Assistant
Princeton University
Sep 2017 – Aug 2021 Princeton, NJ
 
 
 
 
 
Facebook, Inc.
PhD Software Engineer Intern
Facebook, Inc.
Jun 2020 – Aug 2020 Seattle, WA

Projects

Tolerating Slowdowns in Replicated State Machines
Khiem Ngo, Siddhartha Sen, Wyatt Lloyd
Reading Quickly from Storage Across Datacenters
Khiem Ngo, Haonan Lu, Wyatt Lloyd
Supporting Zoomable Video Streams
Khiem Ngo, Guntur Ravindra, Axel Carlier, Mukesh Saini, Cong Pang, Wei Tsang Ooi

Featured Publications

Khiem Ngo, Siddhartha Sen, Wyatt Lloyd
November 2020 OSDI
Tolerating slowdowns in replicated state machines using Copilots

Replicated state machines are linearizable, fault-tolerant groups of replicas that are coordinated using a consensus algorithm. Copilot replication is the first 1-slowdown-tolerant consensus protocol: it delivers normal latency despite the slowdown of any 1 replica.
PDF Code Project Slides Video TR

Recent Publications

Quickly discover relevant content by filtering publications.
Khiem Ngo. Tolerating slowdowns in replicated state machines. PhD Dissertation, Princeton University, 2021.
PDF Project
Khiem Ngo, Haonan Lu, Wyatt Lloyd. K2: Reading quickly from storage across many datacenters. DSN, 2021.
PDF Project Slides
Haonan Lu, Christopher Hodsdon, Khiem Ngo, Shuai Mu, Wyatt Lloyd. The SNOW theorem and latency-optimal read-only transactions. OSDI, 2016.
PDF Project
See all publications

Contact

  • khiem AT khiemn DOT com