{"id":697543,"date":"2020-10-14T12:13:21","date_gmt":"2020-10-14T19:13:21","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-event&p=697543"},"modified":"2025-08-06T11:52:22","modified_gmt":"2025-08-06T18:52:22","slug":"osdi-2020","status":"publish","type":"msr-event","link":"https:\/\/www.microsoft.com\/en-us\/research\/event\/osdi-2020\/","title":{"rendered":"Microsoft at OSDI 2020"},"content":{"rendered":"\n\n

Website:<\/strong> OSDI 2020 (opens in new tab)<\/span><\/a>Opens in a new tab<\/span><\/p>\n

Microsoft is proud to be a\u00a0silver\u00a0sponsor\u00a0of\u00a0the\u00a014th USENIX Symposium on Operating Systems Design and Implementation (opens in new tab)<\/span><\/a>. See more details on our contributions on the sessions tab.<\/p>\n

<\/div>\n

Live chat with us<\/h2>\n

Join us for \u201cAsk Me Anything\u201d(AMA) sessions in the Microsoft Slack Channel during these times:<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Wednesday, November 4<\/strong><\/td>\n<\/tr>\n
Time (PST)<\/strong><\/td>\nName<\/strong><\/td>\n<\/tr>\n
4:00 PM\u20135:00 PM<\/td>\nDan Ports
\nIrene Zhang
\nLidong Zhou<\/td>\n<\/tr>\n
Thursday, November 5<\/strong><\/td>\n<\/tr>\n
Time (PST)<\/strong><\/td>\nName<\/strong><\/td>\n<\/tr>\n
6:00 AM\u20137:00 AM<\/td>\nDushyanth Narayanan<\/td>\n<\/tr>\n
04:00 PM\u201305:00 PM<\/td>\nLidong Zhou
\nRicardo Bianchini<\/td>\n<\/tr>\n
Friday, November 6<\/strong><\/td>\n<\/tr>\n
Time (PST)<\/strong><\/td>\nName<\/strong><\/td>\n<\/tr>\n
06:00 AM\u201307:00 AM<\/td>\nYingnong Dang
\nAleksandar Dragojevic<\/td>\n<\/tr>\n
03:30 PM\u201304:30 PM<\/td>\nAbhisek Pan
\nIshai Menache
\nSebastian Burckhardt<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
<\/div>\n

Program Committee members<\/h3>\n

Andrew Baumann<\/a>
\nChris Hawblitzel<\/a>
\nJay Lorch<\/a>
\nMadan Musuvathi<\/a>
\nDan Ports<\/a>
\nIrene Zhang<\/a>
\nLidong Zhou<\/a>Opens in a new tab<\/span><\/p>\n

Wednesday, November 4<\/h2>\n

7:15 \u2013 8:45 | Correctness<\/h3>\n

Storage Systems are Distributed Systems (So Verify Them That Way!) <\/strong>
\nTravis Hance, Andrea Lattuada, Chris Hawblitzel<\/a>, Jon Howell, Rob Johnson, Bryan Parno<\/p>\n

9:15 \u2013 10:45 | Storage<\/h3>\n

From WiscKey to Bourbon: A Learned Index for Log-Structured Merge Trees <\/strong>
\nYifan Dai, Yien Xu, Aishwarya Ganesan, Ramnatthan Alagappan, Brian Kroth (opens in new tab)<\/span><\/a>, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau<\/p>\n

Generalized Sub-Query Fusion for Eliminating Redundant I\/O from Big-Data Queries <\/strong>
\nPartho Sarthi (opens in new tab)<\/span><\/a>, Kaushik Rajan<\/a>, Akash Lal<\/a>, Abhishek Modi (opens in new tab)<\/span><\/a>, Prakhar Jain (opens in new tab)<\/span><\/a>, Mo Liu (opens in new tab)<\/span><\/a>, Ashit Gosalia (opens in new tab)<\/span><\/a>, Saurabh Kalikar<\/p>\n

1:30 \u2013 3:00 | Consistency<\/h3>\n

Performance-Optimal Read-Only Transactions <\/strong>
\nHaonan Lu (opens in new tab)<\/span><\/a>, Wyatt Lloyd, Siddhartha Sen<\/a><\/p>\n

Pegasus: Tolerating Skewed Workloads in Distributed Storage with In-Network Coherence Directories <\/strong>
\nJialin Li, Jacob Nelson<\/a>, Ellis Michael, Xin Jin, Dan Ports<\/a><\/p>\n

\n

Thursday, November 5<\/h2>\n

7:00 \u2013 8:30 | Machine Learning 1<\/h3>\n

Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads <\/strong>
\nDeepak Narayanan, Keshav Santhanam, Fiodar Kazhamiaka, Amar Phanishayee<\/a>, Matei Zaharia<\/p>\n

HiveD: Sharing a GPU Cluster for Deep Learning with Guarantees <\/strong>
\nHanyu Zhao, Zhenhua Han, Zhi Yang, Quanlu Zhang<\/a>, Fan Yang<\/a>, Lidong Zhou<\/a>, Mao Yang<\/a>, Francis C.M. Lau, Yuqi Wang (opens in new tab)<\/span><\/a>, Yifan Xiong (opens in new tab)<\/span><\/a>, Bin Wang<\/b><\/p>\n

9:00 \u201310:30 | Consensus<\/h3>\n

Blockene: A High-throughput Blockchain Over Mobile Devices<\/strong>
\nSambhav Satija (opens in new tab)<\/span><\/a>, Apurv Mehra<\/a>, Sudheesh Singanamalla, Karan Grover (opens in new tab)<\/span><\/a>, Muthian Sivathanu<\/a>, Nishanth Chandran<\/a>, Divya Gupta<\/a>, Satya Lokam<\/a><\/p>\n

Tolerating Slowdowns in Replicated State Machines using Copilots <\/strong>
\nKhiem Ngo, Siddhartha Sen<\/a>, Wyatt Lloyd<\/p>\n

Byzantine ordered consensus without Byzantine oligarchy <\/strong>
\nYunhao Zhang, Srinath Setty<\/a>, Qi Chen<\/a>, Lidong Zhou<\/a>, Lorenzo Alvisi<\/p>\n

11:15 \u2013 12:45 | Bugs<\/h3>\n

Scalable Runtime Verification of Distributed Middleboxes <\/strong>
\nNofel Yaseen, Behnaz Arzani<\/a>, Ryan Beckett<\/a>, Selim Ciraci (opens in new tab)<\/span><\/a>, Vincent Liu<\/p>\n

1:15 \u2013 2:45 | Scheduling<\/h3>\n

Providing SLOs for Resource-Harvesting VMs in Cloud Platforms <\/strong>
\nLurdh Pradeep Reddy Ambati, Inigo Goiri<\/a>, Felipe Vieira Frujeri<\/a>, Alper Gun, Ke Wang, Brian Dolan (opens in new tab)<\/span><\/a>, Brian Corell (opens in new tab)<\/span><\/a>, Sekhar Pasupuleti (opens in new tab)<\/span><\/a>, Thomas Moscibroda<\/a>, Sameh Elnikety<\/a>, Marcus Fontoura (opens in new tab)<\/span><\/a>, Ricardo Bianchini<\/a><\/p>\n

Protean: VM Allocation Service at Scale <\/strong>
\nOri Hadary<\/a>, Luke Marshall<\/a>, Ishai Menache<\/a>, Abhisek Pan (opens in new tab)<\/span><\/a>, Esaias E Greeff<\/b>, David Dion (opens in new tab)<\/span><\/a>, Star <\/b>Dorminey<\/b>, Shailesh Joshi (opens in new tab)<\/span><\/a>, Yang Chen<\/a>, Mark Russinovich (opens in new tab)<\/span><\/a>, Thomas Moscibroda<\/a><\/p>\n

\n

Friday, November 6<\/h2>\n

7:00 \u2013 8:15 | Machine Learning 2<\/h3>\n

Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks <\/strong>
\nLingxiao Ma (opens in new tab)<\/span><\/a>, Zhiqiang Xie (opens in new tab)<\/span><\/a>, Zhi Yang, Jilong Xue<\/a>, Youshan Miao<\/a>, Wei Cui<\/a>, Wenxiang Hu (opens in new tab)<\/span><\/a>, Fan Yang<\/a>, Lintao Zhang<\/a>, Lidong Zhou<\/a><\/p>\n

A Tensor Compiler for Unified Machine Learning Prediction Serving
\n<\/strong> Supun Nakandala, Karla Saur (opens in new tab)<\/span><\/a>, Gyeong-In Yu, Konstantinos Karanasos<\/a>, Carlo Curino<\/a>, Markus Weimer<\/a>, Matteo Interlandi (opens in new tab)<\/span><\/a><\/p>\n

Retiarii: A Deep Learning Exploratory-Training Framework <\/strong>
\nQuanlu Zhang<\/a>, Zhenhua Han (opens in new tab)<\/span><\/a>, Fan Yang<\/a>, Yuge Zhang (opens in new tab)<\/span><\/a>, Zhe Liu (opens in new tab)<\/span><\/a>, Mao Yang<\/a>, Lidong Zhou<\/a><\/p>\n

8:45 \u2013 10:15 | Hardware<\/h3>\n

Persistent State Machines for Recoverable In-memory Storage Systems with NVRam <\/strong>
\nWen Zhang, Scott Shenker, Irene Zhang<\/a><\/p>\n

12:45 \u2013 2:15 | Clusters<\/h3>\n

Predictive and Adaptive Failure Mitigation to Avert Production Cloud VM Interruptions <\/strong>
\nSebastien Levy (opens in new tab)<\/span><\/a>, Randolph Yao (opens in new tab)<\/span><\/a>, Youjiang Wu (opens in new tab)<\/span><\/a>, Yingnong Dang (opens in new tab)<\/span><\/a>, Peng Huang, Zheng Mu (opens in new tab)<\/span><\/a>, Pu Zhao<\/a>, Tarun Ramani (opens in new tab)<\/span><\/a>, Naga Govindraju (opens in new tab)<\/span><\/a>, Xukun Li (opens in new tab)<\/span><\/a>, Qingwei Lin<\/a>, Gil Lapid Shafriri (opens in new tab)<\/span><\/a>, Murali Chintalapati (opens in new tab)<\/span><\/a><\/p>\n

Unearthing inter-job dependencies for better cluster scheduling <\/strong>
\nAndrew Chung, Subru Krishnan (opens in new tab)<\/span><\/a>, Konstantinos Karanasos<\/a>, Carlo Curino<\/a>, Gregory R. GangerOpens in a new tab<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

Microsoft is proud to be a\u00a0silver\u00a0sponsor\u00a0of\u00a0the\u00a014th USENIX Symposium on Operating Systems Design and Implementation. See details on our contributions to the event this year.<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_startdate":"2020-11-04","msr_enddate":"2020-11-06","msr_location":"Virtual\/Online","msr_expirationdate":"","msr_event_recording_link":"","msr_event_link":"","msr_event_link_redirect":false,"msr_event_time":"","msr_hide_region":false,"msr_private_event":false,"msr_hide_image_in_river":0,"footnotes":""},"research-area":[13547],"msr-region":[256048],"msr-event-type":[197941],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-697543","msr-event","type-msr-event","status-publish","hentry","msr-research-area-systems-and-networking","msr-region-global","msr-event-type-conferences","msr-locale-en_us"],"msr_about":"\n\n

Website:<\/strong> OSDI 2020 (opens in new tab)<\/span><\/a>Opens in a new tab<\/span><\/p>\n

Microsoft is proud to be a\u00a0silver\u00a0sponsor\u00a0of\u00a0the\u00a014th USENIX Symposium on Operating Systems Design and Implementation<\/a>. See more details on our contributions on the sessions tab.<\/p>\n

<\/div>\n

Live chat with us<\/h2>\n

Join us for \u201cAsk Me Anything\u201d(AMA) sessions in the Microsoft Slack Channel during these times:<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Wednesday, November 4<\/strong><\/td>\n<\/tr>\n
Time (PST)<\/strong><\/td>\nName<\/strong><\/td>\n<\/tr>\n
4:00 PM\u20135:00 PM<\/td>\nDan Ports
\nIrene Zhang
\nLidong Zhou<\/td>\n<\/tr>\n
Thursday, November 5<\/strong><\/td>\n<\/tr>\n
Time (PST)<\/strong><\/td>\nName<\/strong><\/td>\n<\/tr>\n
6:00 AM\u20137:00 AM<\/td>\nDushyanth Narayanan<\/td>\n<\/tr>\n
04:00 PM\u201305:00 PM<\/td>\nLidong Zhou
\nRicardo Bianchini<\/td>\n<\/tr>\n
Friday, November 6<\/strong><\/td>\n<\/tr>\n
Time (PST)<\/strong><\/td>\nName<\/strong><\/td>\n<\/tr>\n
06:00 AM\u201307:00 AM<\/td>\nYingnong Dang
\nAleksandar Dragojevic<\/td>\n<\/tr>\n
03:30 PM\u201304:30 PM<\/td>\nAbhisek Pan
\nIshai Menache
\nSebastian Burckhardt<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
<\/div>\n

Program Committee members<\/h3>\n

Andrew Baumann<\/a>
\nChris Hawblitzel<\/a>
\nJay Lorch<\/a>
\nMadan Musuvathi<\/a>
\nDan Ports<\/a>
\nIrene Zhang<\/a>
\nLidong Zhou<\/a>Opens in a new tab<\/span><\/p>\n

Wednesday, November 4<\/h2>\n

7:15 \u2013 8:45 | Correctness<\/h3>\n

Storage Systems are Distributed Systems (So Verify Them That Way!) <\/strong>
\nTravis Hance, Andrea Lattuada, Chris Hawblitzel<\/a>, Jon Howell, Rob Johnson, Bryan Parno<\/p>\n

9:15 \u2013 10:45 | Storage<\/h3>\n

From WiscKey to Bourbon: A Learned Index for Log-Structured Merge Trees <\/strong>
\nYifan Dai, Yien Xu, Aishwarya Ganesan, Ramnatthan Alagappan, Brian Kroth<\/a>, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau<\/p>\n

Generalized Sub-Query Fusion for Eliminating Redundant I\/O from Big-Data Queries <\/strong>
\nPartho Sarthi<\/a>, Kaushik Rajan<\/a>, Akash Lal<\/a>, Abhishek Modi<\/a>, Prakhar Jain<\/a>, Mo Liu<\/a>, Ashit Gosalia<\/a>, Saurabh Kalikar<\/p>\n

1:30 \u2013 3:00 | Consistency<\/h3>\n

Performance-Optimal Read-Only Transactions <\/strong>
\nHaonan Lu<\/a>, Wyatt Lloyd, Siddhartha Sen<\/a><\/p>\n

Pegasus: Tolerating Skewed Workloads in Distributed Storage with In-Network Coherence Directories <\/strong>
\nJialin Li, Jacob Nelson<\/a>, Ellis Michael, Xin Jin, Dan Ports<\/a><\/p>\n

\n

Thursday, November 5<\/h2>\n

7:00 \u2013 8:30 | Machine Learning 1<\/h3>\n

Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads <\/strong>
\nDeepak Narayanan, Keshav Santhanam, Fiodar Kazhamiaka, Amar Phanishayee<\/a>, Matei Zaharia<\/p>\n

HiveD: Sharing a GPU Cluster for Deep Learning with Guarantees <\/strong>
\nHanyu Zhao, Zhenhua Han, Zhi Yang, Quanlu Zhang<\/a>, Fan Yang<\/a>, Lidong Zhou<\/a>, Mao Yang<\/a>, Francis C.M. Lau, Yuqi Wang<\/a>, Yifan Xiong<\/a>, Bin Wang<\/b><\/p>\n

9:00 \u201310:30 | Consensus<\/h3>\n

Blockene: A High-throughput Blockchain Over Mobile Devices<\/strong>
\nSambhav Satija<\/a>, Apurv Mehra<\/a>, Sudheesh Singanamalla, Karan Grover<\/a>, Muthian Sivathanu<\/a>, Nishanth Chandran<\/a>, Divya Gupta<\/a>, Satya Lokam<\/a><\/p>\n

Tolerating Slowdowns in Replicated State Machines using Copilots <\/strong>
\nKhiem Ngo, Siddhartha Sen<\/a>, Wyatt Lloyd<\/p>\n

Byzantine ordered consensus without Byzantine oligarchy <\/strong>
\nYunhao Zhang, Srinath Setty<\/a>, Qi Chen<\/a>, Lidong Zhou<\/a>, Lorenzo Alvisi<\/p>\n

11:15 \u2013 12:45 | Bugs<\/h3>\n

Scalable Runtime Verification of Distributed Middleboxes <\/strong>
\nNofel Yaseen, Behnaz Arzani<\/a>, Ryan Beckett<\/a>, Selim Ciraci<\/a>, Vincent Liu<\/p>\n

1:15 \u2013 2:45 | Scheduling<\/h3>\n

Providing SLOs for Resource-Harvesting VMs in Cloud Platforms <\/strong>
\nLurdh Pradeep Reddy Ambati, Inigo Goiri<\/a>, Felipe Vieira Frujeri<\/a>, Alper Gun, Ke Wang, Brian Dolan<\/a>, Brian Corell<\/a>, Sekhar Pasupuleti<\/a>, Thomas Moscibroda<\/a>, Sameh Elnikety<\/a>, Marcus Fontoura<\/a>, Ricardo Bianchini<\/a><\/p>\n

Protean: VM Allocation Service at Scale <\/strong>
\nOri Hadary<\/a>, Luke Marshall<\/a>, Ishai Menache<\/a>, Abhisek Pan<\/a>, Esaias E Greeff<\/b>, David Dion<\/a>, Star <\/b>Dorminey<\/b>, Shailesh Joshi<\/a>, Yang Chen<\/a>, Mark Russinovich<\/a>, Thomas Moscibroda<\/a><\/p>\n

\n

Friday, November 6<\/h2>\n

7:00 \u2013 8:15 | Machine Learning 2<\/h3>\n

Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks <\/strong>
\nLingxiao Ma<\/a>, Zhiqiang Xie<\/a>, Zhi Yang, Jilong Xue<\/a>, Youshan Miao<\/a>, Wei Cui<\/a>, Wenxiang Hu<\/a>, Fan Yang<\/a>, Lintao Zhang<\/a>, Lidong Zhou<\/a><\/p>\n

A Tensor Compiler for Unified Machine Learning Prediction Serving
\n<\/strong> Supun Nakandala, Karla Saur<\/a>, Gyeong-In Yu, Konstantinos Karanasos<\/a>, Carlo Curino<\/a>, Markus Weimer<\/a>, Matteo Interlandi<\/a><\/p>\n

Retiarii: A Deep Learning Exploratory-Training Framework <\/strong>
\nQuanlu Zhang<\/a>, Zhenhua Han<\/a>, Fan Yang<\/a>, Yuge Zhang<\/a>, Zhe Liu<\/a>, Mao Yang<\/a>, Lidong Zhou<\/a><\/p>\n

8:45 \u2013 10:15 | Hardware<\/h3>\n

Persistent State Machines for Recoverable In-memory Storage Systems with NVRam <\/strong>
\nWen Zhang, Scott Shenker, Irene Zhang<\/a><\/p>\n

12:45 \u2013 2:15 | Clusters<\/h3>\n

Predictive and Adaptive Failure Mitigation to Avert Production Cloud VM Interruptions <\/strong>
\nSebastien Levy<\/a>, Randolph Yao<\/a>, Youjiang Wu<\/a>, Yingnong Dang<\/a>, Peng Huang, Zheng Mu<\/a>, Pu Zhao<\/a>, Tarun Ramani<\/a>, Naga Govindraju<\/a>, Xukun Li<\/a>, Qingwei Lin<\/a>, Gil Lapid Shafriri<\/a>, Murali Chintalapati<\/a><\/p>\n

Unearthing inter-job dependencies for better cluster scheduling <\/strong>
\nAndrew Chung, Subru Krishnan<\/a>, Konstantinos Karanasos<\/a>, Carlo Curino<\/a>, Gregory R. GangerOpens in a new tab<\/span><\/p>\n","tab-content":[{"id":0,"name":"About","content":"Microsoft is proud to be a\u00a0silver\u00a0sponsor\u00a0of\u00a0the\u00a014th USENIX Symposium on Operating Systems Design and Implementation<\/a>. See more details on our contributions on the sessions tab.\r\n

<\/div>\r\n

Live chat with us<\/h2>\r\nJoin us for \u201cAsk Me Anything\u201d(AMA) sessions in the Microsoft Slack Channel during these times:\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
Wednesday, November 4<\/strong><\/td>\r\n<\/tr>\r\n
Time (PST)<\/strong><\/td>\r\nName<\/strong><\/td>\r\n<\/tr>\r\n
4:00 PM\u20135:00 PM<\/td>\r\nDan Ports\r\nIrene Zhang\r\nLidong Zhou<\/td>\r\n<\/tr>\r\n
Thursday, November 5<\/strong><\/td>\r\n<\/tr>\r\n
Time (PST)<\/strong><\/td>\r\nName<\/strong><\/td>\r\n<\/tr>\r\n
6:00 AM\u20137:00 AM<\/td>\r\nDushyanth Narayanan<\/td>\r\n<\/tr>\r\n
04:00 PM\u201305:00 PM<\/td>\r\nLidong Zhou\r\nRicardo Bianchini<\/td>\r\n<\/tr>\r\n
Friday, November 6<\/strong><\/td>\r\n<\/tr>\r\n
Time (PST)<\/strong><\/td>\r\nName<\/strong><\/td>\r\n<\/tr>\r\n
06:00 AM\u201307:00 AM<\/td>\r\nYingnong Dang\r\nAleksandar Dragojevic<\/td>\r\n<\/tr>\r\n
03:30 PM\u201304:30 PM<\/td>\r\nAbhisek Pan\r\nIshai Menache\r\nSebastian Burckhardt<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n
<\/div>\r\n

Program Committee members<\/h3>\r\nAndrew Baumann<\/a>\r\nChris Hawblitzel<\/a>\r\nJay Lorch<\/a>\r\nMadan Musuvathi<\/a>\r\nDan Ports<\/a>\r\nIrene Zhang<\/a>\r\nLidong Zhou<\/a>"},{"id":1,"name":"Sessions","content":"

Wednesday, November 4<\/h2>\r\n

7:15 \u2013 8:45 | Correctness<\/h3>\r\nStorage Systems are Distributed Systems (So Verify Them That Way!) <\/strong>\r\nTravis Hance, Andrea Lattuada, Chris Hawblitzel<\/a>, Jon Howell, Rob Johnson, Bryan Parno\r\n

9:15 \u2013 10:45 | Storage<\/h3>\r\nFrom WiscKey to Bourbon: A Learned Index for Log-Structured Merge Trees <\/strong>\r\nYifan Dai, Yien Xu, Aishwarya Ganesan, Ramnatthan Alagappan, Brian Kroth<\/a>, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau\r\n\r\nGeneralized Sub-Query Fusion for Eliminating Redundant I\/O from Big-Data Queries <\/strong>\r\nPartho Sarthi<\/a>, Kaushik Rajan<\/a>, Akash Lal<\/a>, Abhishek Modi<\/a>, Prakhar Jain<\/a>, Mo Liu<\/a>, Ashit Gosalia<\/a>, Saurabh Kalikar\r\n

1:30 \u2013 3:00 | Consistency<\/h3>\r\nPerformance-Optimal Read-Only Transactions <\/strong>\r\nHaonan Lu<\/a>, Wyatt Lloyd, Siddhartha Sen<\/a>\r\n\r\nPegasus: Tolerating Skewed Workloads in Distributed Storage with In-Network Coherence Directories <\/strong>\r\nJialin Li, Jacob Nelson<\/a>, Ellis Michael, Xin Jin, Dan Ports<\/a>\r\n\r\n
\r\n\r\n

Thursday, November 5<\/h2>\r\n

7:00 \u2013 8:30 | Machine Learning 1<\/h3>\r\nHeterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads <\/strong>\r\nDeepak Narayanan, Keshav Santhanam, Fiodar Kazhamiaka, Amar Phanishayee<\/a>, Matei Zaharia\r\n\r\nHiveD: Sharing a GPU Cluster for Deep Learning with Guarantees <\/strong>\r\nHanyu Zhao, Zhenhua Han, Zhi Yang, Quanlu Zhang<\/a>, Fan Yang<\/a>, Lidong Zhou<\/a>, Mao Yang<\/a>, Francis C.M. Lau, Yuqi Wang<\/a>, Yifan Xiong<\/a>, Bin Wang<\/b>\r\n

9:00 \u201310:30 | Consensus<\/h3>\r\nBlockene: A High-throughput Blockchain Over Mobile Devices<\/strong>\r\nSambhav Satija<\/a>, Apurv Mehra<\/a>, Sudheesh Singanamalla, Karan Grover<\/a>, Muthian Sivathanu<\/a>, Nishanth Chandran<\/a>, Divya Gupta<\/a>, Satya Lokam<\/a>\r\n\r\nTolerating Slowdowns in Replicated State Machines using Copilots <\/strong>\r\nKhiem Ngo, Siddhartha Sen<\/a>, Wyatt Lloyd\r\n\r\nByzantine ordered consensus without Byzantine oligarchy <\/strong>\r\nYunhao Zhang, Srinath Setty<\/a>, Qi Chen<\/a>, Lidong Zhou<\/a>, Lorenzo Alvisi\r\n

11:15 \u2013 12:45 | Bugs<\/h3>\r\nScalable Runtime Verification of Distributed Middleboxes <\/strong>\r\nNofel Yaseen, Behnaz Arzani<\/a>, Ryan Beckett<\/a>, Selim Ciraci<\/a>, Vincent Liu\r\n

1:15 \u2013 2:45 | Scheduling<\/h3>\r\nProviding SLOs for Resource-Harvesting VMs in Cloud Platforms <\/strong>\r\nLurdh Pradeep Reddy Ambati, Inigo Goiri<\/a>, Felipe Vieira Frujeri<\/a>, Alper Gun, Ke Wang, Brian Dolan<\/a>, Brian Corell<\/a>, Sekhar Pasupuleti<\/a>, Thomas Moscibroda<\/a>, Sameh Elnikety<\/a>, Marcus Fontoura<\/a>, Ricardo Bianchini<\/a>\r\n\r\nProtean: VM Allocation Service at Scale <\/strong>\r\nOri Hadary<\/a>, Luke Marshall<\/a>, Ishai Menache<\/a>, Abhisek Pan<\/a>, Esaias E Greeff<\/b>, David Dion<\/a>, Star <\/b>Dorminey<\/b>, Shailesh Joshi<\/a>, Yang Chen<\/a>, Mark Russinovich<\/a>, Thomas Moscibroda<\/a>\r\n\r\n
\r\n\r\n

Friday, November 6<\/h2>\r\n

7:00 \u2013 8:15 | Machine Learning 2<\/h3>\r\nRammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks <\/strong>\r\nLingxiao Ma<\/a>, Zhiqiang Xie<\/a>, Zhi Yang, Jilong Xue<\/a>, Youshan Miao<\/a>, Wei Cui<\/a>, Wenxiang Hu<\/a>, Fan Yang<\/a>, Lintao Zhang<\/a>, Lidong Zhou<\/a>\r\n\r\nA Tensor Compiler for Unified Machine Learning Prediction Serving\r\n<\/strong> Supun Nakandala, Karla Saur<\/a>, Gyeong-In Yu, Konstantinos Karanasos<\/a>, Carlo Curino<\/a>, Markus Weimer<\/a>, Matteo Interlandi<\/a>\r\n\r\nRetiarii: A Deep Learning Exploratory-Training Framework <\/strong>\r\nQuanlu Zhang<\/a>, Zhenhua Han<\/a>, Fan Yang<\/a>, Yuge Zhang<\/a>, Zhe Liu<\/a>, Mao Yang<\/a>, Lidong Zhou<\/a>\r\n

8:45 \u2013 10:15 | Hardware<\/h3>\r\nPersistent State Machines for Recoverable In-memory Storage Systems with NVRam <\/strong>\r\nWen Zhang, Scott Shenker, Irene Zhang<\/a>\r\n

12:45 \u2013 2:15 | Clusters<\/h3>\r\nPredictive and Adaptive Failure Mitigation to Avert Production Cloud VM Interruptions <\/strong>\r\nSebastien Levy<\/a>, Randolph Yao<\/a>, Youjiang Wu<\/a>, Yingnong Dang<\/a>, Peng Huang, Zheng Mu<\/a>, Pu Zhao<\/a>, Tarun Ramani<\/a>, Naga Govindraju<\/a>, Xukun Li<\/a>, Qingwei Lin<\/a>, Gil Lapid Shafriri<\/a>, Murali Chintalapati<\/a>\r\n\r\nUnearthing inter-job dependencies for better cluster scheduling <\/strong>\r\nAndrew Chung, Subru Krishnan<\/a>, Konstantinos Karanasos<\/a>, Carlo Curino<\/a>, Gregory R. Ganger"}],"msr_startdate":"2020-11-04","msr_enddate":"2020-11-06","msr_event_time":"","msr_location":"Virtual\/Online","msr_event_link":"","msr_event_recording_link":"","msr_startdate_formatted":"November 4, 2020","msr_register_text":"Watch now","msr_cta_link":"","msr_cta_text":"","msr_cta_bi_name":"","featured_image_thumbnail":null,"event_excerpt":"Microsoft is proud to be a\u00a0silver\u00a0sponsor\u00a0of\u00a0the\u00a014th USENIX Symposium on Operating Systems Design and Implementation. See details on our contributions to the event this year.","msr_research_lab":[],"related-researchers":[],"msr_impact_theme":[],"related-academic-programs":[],"related-groups":[],"related-projects":[],"related-opportunities":[],"related-publications":[],"related-videos":[],"related-posts":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/697543","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-event"}],"version-history":[{"count":4,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/697543\/revisions"}],"predecessor-version":[{"id":1146927,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/697543\/revisions\/1146927"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=697543"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=697543"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=697543"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=697543"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=697543"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=697543"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=697543"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=697543"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=697543"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}