Yang Wang Assistant Professor Office: DL689 Department of Computer Science and Engineering The Ohio State University Email: wang dot 7564 at osu dot edu

---

My research interests are in distributed systems, in particular fault tolerance, scalability, and performance optimization.
My CV can be found here.

Research:

• wPerf [paper, source code]: this work builds a tool to identify waiting events that are limiting the maximal throughput of a multi-threaded application. To achieve this goal, wPerf first computes how a waiting event can affect threads directly waiting for this event; then wPerf builds a wait-for graph to compute whether such impact can indirectly reach other threads. By combining these two techniques, wPerf essentially tries to identify events with large impacts on all threads.
• SafeTimer [paper, source code]: this work enhances existing timeout detection protocols to tolerate long delays in the OS and the application. At the heartbeat receiver, SafeTimer checks whether there are any pending heartbeats before reporting a failure; at the heartbeat sender, SafeTimer blocks the sender if it cannot send out heartbeats in time. We have proved that SafeTimer can prevent false failure report despite arbitrary delays in the OS and the application. This property allows existing protocols to relax their timing assumptions and use a shorter timeout interval for faster failure detection.
• Hadoop metadata benchmark [paper, source code]: this work builds benchmarks to test HDFS NameNode and Yarn Resource Manager by running real experiments in a small testbed, collecting the traces, and extrapolating such traces to a larger scale.
• ThriftyPaxos [paper, source code]: standard Paxos needs 2f+1 replicas to tolerate f+1 failures. To reduce cost, ThriftyPaxos activates f+1 replicas first and activates backup ones when active ones fail. To ensure system availability when copying data to the newly activated replica, ThriftyPaxos logically separates agreement and execution and exploits a unique property from each: agreement only needs to decide what is the next request, which allows a blank agreement node to join the protocol instantly; execution requires only 1 node to reply when processing a client's request, but requires f+1 nodes to reply when take a snapshot, which means when we have fewer than f+1 replicas, we can still process clients' requests and only need to delay the time-insensitive snapshot task.

Publications:

• "Dayu: Fast and Low-interference Data Recovery in Very-large Storage Systems", Zhufan Wang, Guangyan Zhang, Yang Wang, Qinglin Yang, and Jiaji Zhu, to appear in USENIX ATC 2019.
• "PostMan: Rapidly Mitigating Bursty Traffic by Offloading Packet Processing", Panpan Jin, Jian Guo, Yikai Xiao, Rong Shi, Yipei Niu, Fangming Liu, Chen Qian, and Yang Wang, to appear in USENIX ATC 2019.
• "Henosis: Workload-driven small array consolidation and placement for HDF5 applications on heterogeneous data stores", Donghe Kang, Vadang Patel, Ashwati Nair, Spyros Blanas, Yang Wang, and Srinivasan Parthasarath, to appear in the 33rd ACM International Conference on Supercomputing (ICS '19).
• "Redio: Accelerating Disk-based Graph Processing by Reducing Disk I/Os", Chengwen Wu, Guangyan Zhang, Yang Wang, Xinyang Jiang, and Weimin Zheng, IEEE Transactions on Computers, Oct 2018.
• "wPerf: Generic Off-CPU Analysis to Identify Critical Waiting Events", Fang Zhou, Yifan Gan, Sixiang Ma, and Yang Wang, in OSDI 2018.
• "Evaluating Scalability Bottlenecks by Workload Extrapolation", Rong Shi, Yifan Gan, and Yang Wang, in MASCOTS 2018.
• "Accurate Timeout Detection Despite Arbitrary Processing Delays", Sixiang Ma, and Yang Wang, in USENIX ATC 2018.
• "Cheap and Available State Machine Replication", Rong Shi, and Yang Wang, in USENIX ATC 2016.
• "High-Performance ACID via Modular Concurrency Control". Chao Xie, Chunzhi Su, Cody Littley, Lorenzo Alvisi, Manos Kapritsos, and Yang Wang, in SOSP 2015.
• "Salt: Combining ACID and BASE in a Distributed Database". Chao Xie, Chunzhi Su, Manos Kapritsos, Yang Wang, Navid Yaghmazadeh, Lorenzo Alvisi, and Prince Mahajan, in OSDI 2014.
• "Lazy Means Smart: Reducing Repair Bandwidth Costs in Erasure-coded Distributed Storage". Mark Silberstein, Lakshmi Ganesh, Yang Wang, L. Alvisi, and Mike Dahlin, in Systor 2014. Best paper award
• "Exalt: Empowering Researchers to Evaluate Large-Scale Storage Systems". Yang Wang, Manos Kapritsos, Lara Schmidt, Lorenzo Alvisi, and Mike Dahlin, in NSDI 2014.
• "Robustness in the Salus scalable block store". Yang Wang, Manos Kapritsos, Zuocheng Ren, Prince Mahajan, Jeevetha Kirubanandam, Lorenzo Alvisi, and Mike Dahlin, in NSDI 2013.
• "All about Eve: Execute-Verify Replication for Multi-Core Servers". Manos Kapritsos, Yang Wang, Vivien Quema, Allen Clement, Lorenzo Alvisi, and Mike Dahlin, in OSDI 2012.
• "Gnothi: Separating Data and Metadata for Efficient and Available Storage Replication". Yang Wang, Lorenzo Alvisi, and Mike Dahlin, in USENIX ATC 2012.
• "SOPA: Selecting the Optimal Policy Adaptively". Yang Wang, Jiwu Shu, Guangyan Zhang, Wei Xue, and Weimin Zheng, ACM transactions on storage, Volume 6 Issue 2, July 2010.
• "UpRight Cluster Services". Allen Clement, Manos Kapritsos, Sangmin Lee, Yang Wang, Lorenzo Alvisi, Mike Dahlin, and Taylor Riche, in SOSP 2009.

Services:

• 2019: USENIX ATC Extended Review Committee
• 2018: ICDCS
• 2017: SOSP Student Grant Committee
• 2016: SoCC, SysTOR, DCOSS

Students:

• Sixiang Ma (Ph.D.)
• Yifan Gan (Ph.D.)
• Fang Zhou (Ph.D.)
• Rong Shi (Ph.D. 2018): Facebook
• Zijian Hu (Master 2017): Facebook.

Resources:

• Cluster guidline: here
  
• Public testbeds: CloudLab, Chameleon