Conference Program
- All times are shown in Japanese Standard Time (JST, UTC+9, PDT+16, EDT+13, CEST+7).
- All talks including the keynote speeches are given by pre-recorded videos and an interactive Q&A session follows after each talk on Zoom.
Sponsor Information
August 24 (Mon)
8:45 - 9:00 | Opening Remarks
9:00 - 10:00 | Keynote 1
Session Chair: Taesoo Kim (Georgia Tech)
Speaker: Prof. Vijay Chidambaram (UT Austin)
Building Storage Systems for New Applications, New Hardware, and New Metrics
10:00 - 10:30 | Virtual Break
10:30 - 11:50 | Session 1: Memory and Storage Systems
Session Chair: Ming-chang Yang (CUHK)
- Juno Kim, Yun Joon Soh (University of California, San Diego), Joseph Izraelevitz (University of Colorado, Boulder), Jishen Zhao, Steven Swanson (University of California, San Diego)
SubZero: Zero-copy IO for Persistent Main Memory File Systems **Best Paper Award**
- Gongjin Sun, Sang-Woo Jun (University of California, Irvine)
ColumnBurst: A Near-Storage Accelerator for Memory-Efficient Database Join Queries
- Youyun Wang, Chuzhe Tang, Zhaoguo Wang, Haibo Chen (Shanghai Jiao Tong University)
SIndex: A Scalable Learned Index for String Keys
- Ryo Nakamura, Yohei Kuga (The University of Tokyo), Kunio Akashi (NICT)
How beneficial is Peer-to-Peer DMA?
12:00 - 13:30 | Virtual Lunch Break
13:30 - 14:50 | Session 2: Operating Systems
Session Chair: Jean-Pierre Lozi (Oracle Labs)
-
Giorgos Kappes, Stergios Anastasiadis (University of Ioannina)
Libservices: Dynamic Storage Provisioning for Multitenant I/O Isolation
-
Stefan Reif, Benedict Herzog, Fabian Hügel, Timo Hönig, Wolfgang Schröder-Preikschat (Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU))
Nearly Symmetric Multi-Core Processors
-
Matthias Hille (TU Dresden / Huawei Dresden Research Center), Nils Asmussen (Barkhauseninstitut), Hermann Härtig (TU Dresden), Pramod Bhatotia (University of Edinburgh)
Heterogeneous Microkernel OS for Rack-Scale Systems
-
Mathias Gottschlag, Tim Schmidt, and Frank Bellosa (Karlsruhe Institute of Technology)
AVX Overhead Profiling: How Much Does Your Fast Code Slow You Down?
August 25 (Tue)
9:00 - 10:00 | Keynote 2
Session Chair: Patrick P. C. Lee (CUHK)
Speaker: Prof. Xin Jin (Johns Hopkins University)
Building Next-Generation Cloud Systems with Programmable Hardware
10:00 - 10:30 | Virtual Break
10:30 - 11:50 | Session 3: Machine Learning and AI
Session Chair: Patrick P. C. Lee (CUHK)
-
Jingde Chen, Subho S. Banerjee, Zbigniew Kalbarczyk, Ravishankar K. Iyer (University of Illinois at Urbana-Champaign)
Machine Learning for Load Balancing in the Linux Kernel
-
Shiqi Jiang, Ting Cao, Lihao Ran, Yunxin Liu (Microsoft Research Asia)
Profiling and Optimizing Deep Learning Inference on Mobile GPUs
-
Xiaorui Wu, Hong Xu (City University of Hong Kong), Honglin Zhang, Huaming Chen, Jian Wang (Tencent)
Saec: Similarity-Aware Embedding Compression in Recommendation Systems
-
Minsub Kim, Sungjin Lee (DGIST)
Reducing Tail Latency of DNN-based Recommender System using In-Storage Processing
12:00 - 13:30 | Virtual Lunch Break
13:30 - 14:50 | Session 4: Distributed Systems and Networks
Session Chair: Qun Huang (Peking University)
-
Jifei Yi, Benchao Dong, Mingkai Dong, Haibo Chen (Shanghai Jiao Tong University)
On the Precision of Precise Event Based Sampling
-
Rengan Dou, Xin Wang, Richard T. B. Ma (National University of Singapore)
Towards Application Performance Fairness on Clouds
-
Yiran Li, Kevin Gao (Tsinghua University), Xin Jin (Johns Hopkins University), Wei Xu (Tsinghua Univesity)
Concerto: Cooperative Network-Wide Telemetry with Controllable Error Rate
-
Ryan Nathanael Soenjoto Widodo, Hirotake Abe, Kazuhiko Kato (University of Tsukuba)
HDRF: Hadoop Data Reduction Framework for Hadoop Distributed File System