I lead the UT Systems and
Storage Lab. I am also a part of
the LASR research
Research. My group aims to build the next generation of
storage systems. This means building storage systems that have
higher performance and stronger reliability. We build both storage
systems and the tools need to develop such systems rigorously,
such as testing frameworks. Our work involves innovation both at
the data-structure level and at the systems
level. We open-source
all the software we build, and aim to have impact both inside
academia and in industry.
Recent major projects:
- MONet. Framework
for reducing memory consumption of DNN training.
- SplitFS. File
system for persistent memory which reduces software
overhead. SOSP 2019.
- RECIPE. Concurrent,
crash-consistent persistent-memory indexes. SOSP 2019.
and Ace. Tools to systematically test file-system
crash consistency. OSDI 2018
- PebblesDB. Write-optimized
key-value store that outperforms LevelDB and RocksDB. SOSP
- Removing storage bottlenecks in training deep-learning
- Decreasing memory consumption for CNN models (Aashaka Shah)
- Building a high-performance persistent-memory key-value
- Improving performance of persistent-memory applications
and file systems (Rohan
- Building data structures on dis-aggregated persistent
- Finding bugs in persistent memory file systems
- Verifying Linux kernel file systems
- Crashmonkey-generated tests have been added to xfstests,
the file-system test suite for the Linux kernel
- Crashmonkey found 10 long-standing bugs in btrfs and F2FS
in the Linux kernel.
- Crashmonkey found a bug in the FSCQ verified file system
which caused the file system to not persist data on
fdatasync. The bug was in the unverified part of the file
system (Haskell-C bindings).
- The techniques behind PebblesDB are being used in the
storage product of a major IT company.
My research is supported
and Google. I am
Researcher at VMware Research.
Student Recruiting: I do not look at GRE scores
when recruiting students. Letters and research experience are the
biggest factors for me. I generally interview students (remotely), and
that is the strongest signal of whether I would work with them. If you
are applying to UTCS to work with me, you don't need to take the GRE
for Fall 2021 PhD admissions. I am not recruiting MS students.
Dec 2020: My blog post
We Should Include One-Shot Revision in our Review Process" has
appeared on the SIGARCH
blog and SIOPGS blog!
Aug 2020: I will be
co-chairing SYSTOR 2021
with Prof. Danny Raz!
Aug 2020: I'm giving one of the keynote talks at ApSys 20!
Mar 2020: My blog post
Changing World of Storage" has appeared on the SIGARCH blog!
2020: Our work
on GDPRbench has been
accepted to VLDB 2020!
Congratulations to Supreeth and the team!
has been selected as one of the finalists for the "Memorable Paper"
Award at the 11th Annual Non-Volatile
Jan 2020: I will be serving as a contributor to
the SIGARCH blog for 2020!
Jan 2020: Rohan and Se Kwon will be
and RECIPE at the
11th Annual Non-Volatile Memories Workshop!
Dec 2019: I will be co-chairing HotStorage 2020
Badam! Send us your best work!
Nov 2019: Baris
Kasikci and I will serving as editors for
the SIGOPS Blog. Contact us
if you would like to contribute an article!
Recent Publications (All)
Memory Optimization for Deep Networks
Analyzing and Mitigating Data Stalls in DNN Training
CheckFreq: Frequent, Fine-Grained DNN Checkpointing
SplinterDB: Closing the Bandwidth Gap for NVMe Key-Value
Understanding and Benchmarking the Impact of GDPR on Database Systems
Kumar, Vijay Chidambaram
SplitFS: Reducing Software Overhead in File Systems for
Se Kwon Lee,
RECIPE : Converting Concurrent DRAM Indexes to Persistent-Memory Indexes
Full List of Publications
Se Kwon Lee,
Work in Progress
Scalable and Efficient Data Authentication for Decentralized Systems
Shah, Souvik Banerjee, Vijay Chidambaram, Dahlia
Malkhi, Michael Wei
We Should Include One-Shot Revision in our Review Process
The Changing World of Storage
Oct 2020. Will be talking about bootstrapping a research group
at the New Faculty
Symposium at ISSRE 2020!
Aug 2020. Gave one of keynote talks at
ACM SIGOPS Asia-Pacific Workshop on
the video and slides.
Mar 2020. Keynote at The First International Workshop on
Distributed Ubiquitous Computing: Systems, Applications, and
Networking (DUCSAN 2020).
Jan 2020. Presented CrashMonkey at VMware's Tech Talk
Nov 2019. Presented our work on GDPR at Vmware.
Apr 2019. Presented PebblesDB at Micron, Austin.
PC Member for:
SRC (with SOSP),
MASCOTS, MSST, HotStorage,
2016: HotStorage, NVMW
External Reviewer/Review Committee Member
for ASPLOS 20, ASPLOS 18, ASPLOS 17,
Reviewer for TOCS, TACO, TDSC, TOS
NSF Panelist in 2018
- Best Paper Award at ATC 2018
- Best Paper Award at FAST 2018
CAREER Award 2018
- Best Poster Award at ApSys 2017
- Best Paper Award at FAST 2017
- ACM SIGOPS Dennis M. Ritchie Dissertation Award 2016
- Microsoft Research Graduate Fellowship 2014
- University of Wisconsin-Madison Alumni Scholarship 2009
- Jayashree Mohan: Microsoft Research PhD Fellowship 2019
- Ashlie Martinez: CRA Best Undergraduate Researcher 2018
- Ashlie Martinez: UT Austin Dean's Honored Graduate 2018
- Ashlie Martinez: UT Austin Aspire Research Excellence Award 2018
If you are a student considering grad school or applying for
grad school, these links may be
useful (advice on contacting profs, getting letters of
the Systems Slack,
a forum for profs, students, post-docs, and industry to discuss
ideas related to systems research and practice.
I started and help
maintain the Computer
Science Research Tales (CSRTales) blog, where we present the
backstories behind CS research projects.
the SOSP/OSDI Hall of Fame. This page
indicates who has published the most at these premier venues, and
who has been publishing recently. These are good guides for grad
students trying to decide which professor to work with!