Welcome!
Yanpei Chen
My nice half and my evil half : ) Can you tell which is which?
ychen [at] berkeley [dot] edu. Last updated February 2012.
Research Interests
I am interested in Internet datacenters.
Datacenters are the nexus of information that facilitate storage, index, search,
and other core Internet functionalities. They are a microcosm of the wider Internet, presenting
its own design and research challenges. They also represent a highly managed environment
within leading Internet businesses, offering the opportunity for experimentation with
immediate impact. My current interests include MapReduce/Hadoop energy efficiency,
and datacenter workload characterization. Recently I've also developed an interest in
cloud computing security, a new area that's still somewhat blurry - hopefully what little
knowledge I have about datacenters will be helpful.
I am a member of the
AMP Lab
(formerly RAD Lab) at
UC Berkeley,
and my research advisor is
Professor Randy H. Katz.
I have on-going collaborations with RAD Lab industry partners, most recently including
Cloudera, NetApp,
and Facebook.
Recently we released a Statistical Workload Injector for MapReduce (SWIM),
which includes workload synthesis/replay tools and some workloads pre-generated from Facebook production MapReduce traces.
You can learn more about SWIM in our
MASCOTS 2011 paper.
Publications / Talks
- Y. Chen, S. Alspaugh, D. Borthakur, R. Katz. Energy Efficiency for Large-Scale MapReduce Workloads with Significant Interactive Analysis. 2012.
In press. European Conference on Computer Systems (EuroSys) 2012.
- Y. Chen, A. Ganapathi, R. Katz. Challenges and Opportunities for Managing Data Systems Using Statistical Models. 2011.
IEEE Data Engineering Bulletin. Vol. 34. No. 4. pp. 53-60. December 2011.
- Y. Chen, K. Srinivasan, G. Goodson, R. Katz. Design Implications for Enterprise Storage Systems via Multi-Dimensional Trace Analysis. 2011.
23rd ACM Symposium on Operating Systems Principles (SOSP) 2011.
- Y. Chen, A. Ganapathi, R.Griffith, R. Katz. The Case for Evaluating MapReduce Performance Using Workload Suites. 2011.
19th IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS) 2011.
- A. Krioukov, C. Goebel, S. Alspaugh, Y. Chen, D. Culler, R. Katz Integrating Renewable Energy Using Data Analytics Systems: Challenges and Opportunities. 2011.
IEEE Data Engineering Bulletin. Vol. 34. No. 1. pp. 3-15. March 2011.
- Y. Chen, R. Katz. Glimpses of the Brave New World for Cloud Security. 2011.
Feature article. HPC in the Cloud. 2011.
- R. Katz, D. Culler, S. Sanders, S. Alspaugh, Y. Chen, et al. An Information-Centric Energy Infrastructure: The Berkeley View. 2011.
Sustainable Computing: Informatics and Systems. Volume 1. Issue 1. pp. 7-22. 2011.
- Y. Chen, A. Ganapathi, R. Katz. To Compress or Not To Compress - Compute vs. IO tradeoffs for MapReduce Energy Efficiency. 2010.
First ACM SIGCOMM Workshop on Green Networking, co-located with SIGCOMM 2010.
- A. Ganapathi, Y. Chen, A. Fox, R. Katz, D. Patterson. Statistics-Driven Workload Modeling for the Cloud. 2010.
Most Popular Paper Award, Workshop on Self-Managing Database Systems (SMDB) 2010, co-located with ICDE 2010.
- Y. Chen, L. Keys, R. H. Katz. Towards Energy Efficient Hadoop. 2009.
Hadoop Summit 2009, presented by Yahoo!
- Y. Chen, R. Griffith, J. Liu, R. H. Katz, A. D. Joseph. Understanding TCP Incast Throughput Collapse in Datacenter Networks. 2009.
Workshop on Research in Enterprise Networking (WREN) 2009, co-located with SIGCOMM 2009.
- Y. Chen, T. X. Wang, and R. H. Katz. Energy Efficient Ethernet Encodings. 2008.
33rd Annual IEEE Conference on Local Computer Networks (LCN), 2008.
- Y. Chen & D. Garcia et al. 200 Students Can't Be Wrong! GamesCrafters, a Computational Game Theory Undergraduate Research and Development Group. 2007.
2nd AAAI Symposium on Using AI to Motivate Greater Participation in Computer Science.
Other random stuff
Outside of research, I am interested in the overlap between technology and other non-techical traditions.
At various times, I have ventured into high-tech business, cyberlaw, technology policy, technology culture,
scientific history, information and innovation studies, technology education, and other similar areas.
I believe that business, law, politics, culture, policy, and education are all
intimately connected with technology. They affect each other in complex ways, and will become even more mutally dependent in the future.
I think we computer scientists should
participate in the discussion and contribute our unique perspectives. Some half-presentable work from the past:
- Y. Chen & J. Nam. Gender Balance in UC Berkeley EECS. 2007.
Self-started research project. Results released in Spring 2007 to the Chair of EECS and the Diversity Director of the Department. Met with UC Berkeley Chancellor & the Assoc. Vice Provost for Faculty Equity to discuss results. Report forwarded to the Vice Chancellor for Equity and Inclusion, as well as other entities on campus.
- Y. Chen. Morality and Political Discourse. 2005.
Winning entry to the UC Berkeley Leslie Lipson Humanitistic Values Scholarship Contest. Written after the 2004 Presidential Elections to temper the somewhat reactionary criticism of the political atmosphere and to point out new opportunities.
Personal stuff
I was born in China, my whole family is in the United States, and I
am technically an Australian. I hold interests unique to each country - I
do Chinese caligraphy, I love College Football (Go Bears!!!), and I still
itch to play cricket. I went to an all-guys public high school in
Australia, and apparently I still owe the teachers there some homework and
beer and lost bets. I had a wonderful time as undergrad here at UC
Berkeley. I am a graduate student now, but I will always be an
undergrad at heart. My parents are
chemists/chemical engineers; they live and work in New York.
I play computer games a lot (most recently Civilization 4, Rome Total War, and Starcraft 2),
I once was addicted to Friends, I'm reading the Game of Throne series and plan to re-read
100 Years of Solitude (didn't fully "get it" the first time but found it awesome),
and I love Clint Eastwood and Al Pacino movies. I play tennis/badminton/table-tennis,
I've made a home-recorded album for classical guitar, I hike sporadically,
and I've lost $8 playing chess against homeless guys.
Recently I've taken up archery
- still a newbie shooting a 25-pound bow at 30 yards.
I've also taken up fencing - doing obligatory newbie French-Grip foil. I'm currently
focused on trying to not get hit so unstylishly in fencing, but alway looking for the next hobby too.
The rest to be filled later ...
Awards
Current National Science Foundation Graduate Research Fellow
UC Berkeley Regents and Chancellors Scholar
UC Berkeley Lipson Humanistic Values Scholar
Premier's Awards in Math, Physics, Chemistry (Victoria, Australia)
Australia Student Prize 2001