Portrait of Cheng Huang

Cheng Huang

Principal Researcher

About

Cheng Huang is a research scientist and tech lead at the Microsoft Azure Storage team.

  • Responsible for all erasure coding strategies (within DC, within region and cross-region) in Azure Storage.
  • Leads the production development of Azure Storage vNext (to scale the capacity of Azure Storage by more than 100x).
  • Educates Azure teams on formal methods (such as TLA+ and IronFleet) and model checking (such as P#). Organizes the Dr. TLA+ Series.

Projects

Safe asynchronous programming with P and P#

Established: March 15, 2016

We are designing programming languages for building safe and reliable asynchronous systems. The languages are based on the programming idiom of communicating state machines. They offer first-class support for writing safety and liveness specifications as well as building abstract models of…

Cloud Faster

Established: February 25, 2010

To make cloud computing work, we must make applications run substantially faster, both over the Internet and within data centers. Our measurements of real applications show that today's protocols fall short, leading to slow page-load times across the Internet and…

Publications

2016

2014

2012

On the Locality of Codeword Symbols
Parikshit Gopalan, Cheng Huang, Huseyin Simitci, Sergey Yekhanin, in IEEE Transactions on Information theory 2012; preliminary version in Allerton 2011 (Invited paper); winner of the 2014 IEEE Communication and Information Theory Society Joint Paper Prize, IEEE, November 1, 2012, View abstract, Download PDF

2011

2010

2009

2008

2007

2005

2004

Projects

Other

Dr. Huang has also been working on erasure coding technologies for storage applications. While the industry has been relying on conventional schemes, invented more than 50 years ago and primed in deep space communication, Dr. Huang and his MSR colleagues have partnered with the Windows Azure Storage team and designed novel schemes that are optimized specifically for cloud storage. The LRC paper received a Best Paper Award at USENIX ATC 2012. LRC significantly reduces storage cost over conventional schemes, while still maintaining required durability and I/O performance. Given the scale and growth of Azure, the savings from LRC translate into 100s of millions of dollars for Microsoft. This work has been covered by the Microsoft Research featured story.

In the past, Dr. Huang and his MSR colleagues have partnered with the MSN team, designed and developed Microsoft’s peer-assisted Video-on-Demand system, which significantly reduces the cost of video delivery while dramatically boosting video quality at the same time. Please check out the profitable VoD paper at ACM SIGCOMM 2007. This work has also been featured by New Scientist.

Dr. Huang has partnered with Azure and built a Global Traffic Management system; the Apollo paper was presented at USENIX ATC 2010. Dr. Huang has partnered with the Xbox team, designed and built the core decision engine in Pangolin—the low delay network transport protocol used in Xbox social gaming—and presented the Pangolin paper at ACM CoNEXT in 2011. Dr. Huang has also partnered with the Bing team and investigated Internet Search hijacking. This work was presented in the Revealer paper at USENIX LEET 2011 and featured in stories by the Register and Ars Technica.

Dr. Huang received his B.S. and M.S. degrees from Shanghai Jiao Tong University, both in Electronic Engineering, and his Ph.D. degree in Computer Science from Washington University in St. Louis.

 

Professional Services

Technical Program Committee (TPC)