Skip to content

Commit 5a95a99

Browse files
ryanphuangryanphuang
committed
deploy: a8a182e 
1 parent d475aa1 commit 5a95a99

14 files changed

Lines changed: 90 additions & 57 deletions

‎assets/img/team/juntao.jpg‎

156 KB
Loading

‎feed.xml‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.2.2">Jekyll</generator><link href="/feed.xml" rel="self" type="application/atom+xml" /><link href="/" rel="alternate" type="text/html" /><updated>2025-08-24T14:15:24-04:00</updated><id>/feed.xml</id><title type="html">Ordered Systems Lab at U-M</title><subtitle>This is the website for the Ordered Systems Lab (a.k.a Order Lab) at University of Michigan, led by Prof. Ryan Huang.</subtitle></feed>
1+
<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.2.2">Jekyll</generator><link href="/feed.xml" rel="self" type="application/atom+xml" /><link href="/" rel="alternate" type="text/html" /><updated>2025-11-18T13:20:27-05:00</updated><id>/feed.xml</id><title type="html">Ordered Systems Lab at U-M</title><subtitle>This is the website for the Ordered Systems Lab (a.k.a Order Lab) at University of Michigan, led by Prof. Ryan Huang.</subtitle></feed>

‎paper/atropos-sosp25.pdf‎

979 KB
Binary file not shown.

‎paper/atropos.bib‎

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
@inproceedings{AtroposSOSP2025,
2+
author = {Hu, Yigong and Zhang, Zeyin and Liu, Yicheng and Gu, Yile and Lei, Shuangyu and Kasikci, Baris and Huang, Peng},
3+
title = {Mitigating Application Resource Overload with Targeted Task Cancellation},
4+
year = {2025},
5+
isbn = {9798400718700},
6+
publisher = {Association for Computing Machinery},
7+
address = {New York, NY, USA},
8+
url = {https://doi.org/10.1145/3731569.3764835},
9+
doi = {10.1145/3731569.3764835},
10+
booktitle = {Proceedings of the ACM SIGOPS 31st Symposium on Operating Systems Principles},
11+
pages = {270–285},
12+
numpages = {16},
13+
keywords = {overload control, resource contention, request cancellation},
14+
location = {Lotte Hotel World, Seoul, Republic of Korea},
15+
series = {SOSP '25}
16+
}

‎paper/phoenix-sosp25.pdf‎

915 KB
Binary file not shown.

‎paper/phoenix.bib‎

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
@inproceedings{PhoenixSOSP2025,
2+
author = {Jing, Yuzhuo and Mai, Yuqi and Cai, Angting and Chen, Yi and He, Wanning and Qian, Xiaoyang and Chen, Peter M. and Huang, Peng},
3+
title = {Optimistic Recovery for High-Availability Software via Partial Process State Preservation},
4+
year = {2025},
5+
isbn = {9798400718700},
6+
publisher = {Association for Computing Machinery},
7+
address = {New York, NY, USA},
8+
url = {https://doi.org/10.1145/3731569.3764858},
9+
doi = {10.1145/3731569.3764858},
10+
booktitle = {Proceedings of the ACM SIGOPS 31st Symposium on Operating Systems Principles},
11+
pages = {305–321},
12+
numpages = {17},
13+
keywords = {availability, software recovery, operating systems, static analysis},
14+
location = {Lotte Hotel World, Seoul, Republic of Korea},
15+
series = {SOSP '25}
16+
}

‎paper/trainverify-sosp25.pdf‎

1.34 MB
Binary file not shown.

‎paper/trainverify.bib‎

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
@inproceedings{TrainVerifySOSP2025,
2+
author = {Lu, Yunchi and Miao, Youshan and Tan, Cheng and Huang, Peng and Zhu, Yi and Zhang, Xian and Yang, Fan},
3+
title = {{TrainVerify}: Equivalence-Based Verification for Distributed {LLM} Training},
4+
year = {2025},
5+
isbn = {9798400718700},
6+
publisher = {Association for Computing Machinery},
7+
address = {New York, NY, USA},
8+
url = {https://doi.org/10.1145/3731569.3764850},
9+
doi = {10.1145/3731569.3764850},
10+
booktitle = {Proceedings of the ACM SIGOPS 31st Symposium on Operating Systems Principles},
11+
pages = {237–253},
12+
numpages = {17},
13+
keywords = {deep neural network, model parallelization, equivalence checking, formal verification, symbolic execution},
14+
location = {Lotte Hotel World, Seoul, Republic of Korea},
15+
series = {SOSP '25}
16+
}

‎paper/xinda-nsdi25.bib‎

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6,4 +6,8 @@ @inproceedings{SlowFaultStudy2025NSDI
66
month = {April},
77
year = {2025},
88
location = {Philadelphia, PA, USA},
9+
isbn = {978-1-939133-46-5},
10+
pages = {359--378},
11+
url = {https://www.usenix.org/conference/nsdi25/presentation/lu},
12+
publisher = {USENIX Association},
913
}

‎pubs.html‎

Lines changed: 12 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -66,19 +66,22 @@ <h1>Publications</h1>
6666
<h2 id="publications">2025</h2>
6767
<ul class="publications">
6868
<li>
69-
<a target="_blank" href="#">Verifying Distributed Deep Learning Training via Parallelization Equivalence</a><br>
70-
<span class="authorlist"><i><a href="https://mercury-browser-ede.notion.site/yunchi" class="nodec">Yunchi Lu</a>, </i><i><a href="https://naizhengtan.github.io" class="nodec">Cheng Tan</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/yomia" class="nodec">Youshan Miao</a>, </i><i><a href="https://web.eecs.umich.edu/~ryanph" class="nodec">Peng Huang</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/yizhu1" class="nodec">Yi Zhu</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/zhxian" class="nodec">Xian Zhang</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/fanyang" class="nodec">Fan Yang</a><br></i></span>
71-
<a target="_blank" href="https://sigops.org/s/conferences/sosp/2025/" class="conf"><b>SOSP 2025</b></a>
69+
<a target="_blank" href="paper/trainverify-sosp25.pdf">TrainVerify: Equivalence-Based Verification for Distributed LLM Training</a><br>
70+
<span class="authorlist"><i><a href="https://luyunchi.github.io" class="nodec">Yunchi Lu</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/yomia" class="nodec">Youshan Miao</a>, </i><i><a href="https://naizhengtan.github.io" class="nodec">Cheng Tan</a>, </i><i><a href="https://web.eecs.umich.edu/~ryanph" class="nodec">Peng Huang</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/yizhu1" class="nodec">Yi Zhu</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/zhxian" class="nodec">Xian Zhang</a>, </i><i><a href="https://www.microsoft.com/en-us/research/people/fanyang" class="nodec">Fan Yang</a><br></i></span>
71+
<a target="_blank" href="https://sigops.org/s/conferences/sosp/2025/" class="conf"><b>SOSP 2025</b></a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="paper/trainverify.bib">citation</a>
72+
&nbsp;&nbsp;<a target="_blank" role="button" class="btn btn-outline-primary publinkitem" href="slides/trainverify_sosp25_slides.pdf">slides</a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="https://github.com/verify-llm/TrainVerify">software</a>&nbsp;&nbsp;<a target="_blank" role="button" class="btn btn-outline-primary publinkitem" href="https://arxiv.org/abs/2506.15961">arXiv</a>
7273
</li>
7374
<li>
74-
<a target="_blank" href="#">Optimistic Recovery for High-Availability Software via Partial Process State Preservation</a><br>
75+
<a target="_blank" href="paper/phoenix-sosp25.pdf">Optimistic Recovery for High-Availability Software via Partial Process State Preservation</a><br>
7576
<span class="authorlist"><i><a href="https://osdi.dev" class="nodec">Yuzhuo Jing</a>, </i><i>Yuqi Mai, </i><i>Angting Cai, </i><i><a href="https://chenyi.world" class="nodec">Yi Chen</a>, </i><i><a href="https://hwanning.netlify.app" class="nodec">Wanning He</a>, </i><i>Xiaoyang Qian, </i><i><a href="https://web.eecs.umich.edu/~pmchen" class="nodec">Peter M. Chen</a>, </i><i><a href="https://web.eecs.umich.edu/~ryanph" class="nodec">Peng Huang</a><br></i></span>
76-
<a target="_blank" href="https://sigops.org/s/conferences/sosp/2025/" class="conf"><b>SOSP 2025</b></a>
77+
<a target="_blank" href="https://sigops.org/s/conferences/sosp/2025/" class="conf"><b>SOSP 2025</b></a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="paper/phoenix.bib">citation</a>
78+
&nbsp;&nbsp;<a target="_blank" role="button" class="btn btn-outline-primary publinkitem" href="slides/phoenix_sosp25_slides.pdf">slides</a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="https://github.com/OrderLab/phoenix">software</a>
7779
</li>
7880
<li>
79-
<a target="_blank" href="#">Mitigating Application Resource Overload with Targeted Task Cancellation</a><br>
81+
<a target="_blank" href="paper/atropos-sosp25.pdf">Mitigating Application Resource Overload with Targeted Task Cancellation</a><br>
8082
<span class="authorlist"><i><a href="https://yigonghu.github.io" class="nodec">Yigong Hu</a>, </i><i>Zeyin Zhang, </i><i>Yicheng Liu, </i><i>Yile Gu, </i><i>Shuangyu Lei, </i><i><a href="https://homes.cs.washington.edu/~baris" class="nodec">Baris Kasikci</a>, </i><i><a href="https://web.eecs.umich.edu/~ryanph" class="nodec">Peng Huang</a><br></i></span>
81-
<a target="_blank" href="https://sigops.org/s/conferences/sosp/2025/" class="conf"><b>SOSP 2025</b></a>
83+
<a target="_blank" href="https://sigops.org/s/conferences/sosp/2025/" class="conf"><b>SOSP 2025</b></a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="paper/atropos.bib">citation</a>
84+
&nbsp;&nbsp;<a target="_blank" role="button" class="btn btn-outline-primary publinkitem" href="slides/atropos_sosp25_slides.pdf">slides</a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="https://github.com/OrderLab/Atropos">software</a>
8285
</li>
8386
<li>
8487
<a target="_blank" href="paper/traincheck-osdi25-preprint.pdf">Training with Confidence: Catching Silent Errors in Deep Learning Training with Automated Proactive Checks</a><br>
@@ -93,7 +96,7 @@ <h2 id="publications">2025</h2>
9396
</li>
9497
<li>
9598
<a target="_blank" href="paper/xinda-nsdi25-preprint.pdf">One-Size-Fits-None: Understanding and Enhancing Slow-Fault Tolerance in Modern Distributed Systems</a><br>
96-
<span class="authorlist"><i><a href="https://ruiming-lu.github.io" class="nodec">Ruiming Lu</a>, </i><i><a href="https://mercury-browser-ede.notion.site/yunchi" class="nodec">Yunchi Lu</a>, </i><i><a href="https://essoz.github.io" class="nodec">Yuxuan Jiang</a>, </i><i>Guangtao Xue, </i><i><a href="https://web.eecs.umich.edu/~ryanph" class="nodec">Peng Huang</a><br></i></span>
99+
<span class="authorlist"><i><a href="https://ruiming-lu.github.io" class="nodec">Ruiming Lu</a>, </i><i><a href="https://luyunchi.github.io" class="nodec">Yunchi Lu</a>, </i><i><a href="https://essoz.github.io" class="nodec">Yuxuan Jiang</a>, </i><i>Guangtao Xue, </i><i><a href="https://web.eecs.umich.edu/~ryanph" class="nodec">Peng Huang</a><br></i></span>
97100
<a target="_blank" href="https://www.usenix.org/conference/nsdi25" class="conf"><b>NSDI 2025</b></a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="paper/xinda-nsdi25.bib">citation</a>
98101
&nbsp;&nbsp;<a target="_blank" role="button" class="btn btn-outline-primary publinkitem" href="slides/xinda_nsdi25_slides.pdf">slides</a>&nbsp;&nbsp;<a target="_blank" class="btn btn-outline-primary publinkitem" href="https://github.com/OrderLab/xinda">software</a><br><div class="press"><b>Coverage:</b> <a target="_blank" href="https://cse.engin.umich.edu/stories/a-new-tool-to-manage-slow-faults">CSE News</a>, <a target="_blank" href="https://techxplore.com/news/2025-05-tool-faults-real-adjustment.html">Tech Xplore</a> </div>
99102
</li>
@@ -357,15 +360,14 @@ <h2 id="publications">2010</h2>
357360
</div>
358361
</section>
359362

360-
361363
<!-- ================ Start footer area ================= -->
362364
<footer class="footer-area section-gap">
363365
<div class="container">
364366
<div class="footer-bottom row align-items-center text-center text-lg-left">
365367
<p class="footer-text m-0 col-lg-8 col-md-12">
366368
Copyright &copy; OrderLab 2017-<script>
367369
document.write(new Date().getFullYear());
368-
</script> All rights reserved. | Last updated 2025-07-28 14:01:59 -0400.
370+
</script> All rights reserved. | Last updated 2025-11-18 13:19:04 -0500.
369371
</p>
370372
</div>
371373
</div>

0 commit comments

Comments
 (0)