SING Weekly Discussions Archive
This is the PhD List Order1.
Welcome to choose your own topic of ML-Tutorials.
This is the SING-LAB List Order2.
| Date |
Pre senter |
Topic |
Source |
Links |
|
|
|
|
|
|
|
|
|
|
| 10/3 |
Yiding |
INFaaS on model-less machine learning inference |
Tutorial |
[/] |
##September
| Date |
Pre senter |
Topic |
Source |
Links |
| 9/20 |
LiuYang |
advertising recommendation |
Papers |
[/] |
| 9/12 |
Chaoliang |
Distributed machine learning |
Papers |
[/] |
| 9/5 |
Gechen |
Clutering |
Tutorial |
[/] |
| Date |
Presenter |
Topic |
Source |
Links |
| 8/22 |
Hao |
Flow scheduling (4) |
tutorial |
[/] |
| 8/15 |
Xinchen |
ML Reinforcement Learning (3) |
tutorial |
[/] |
| Date |
Presenter |
Topic |
Source |
Links |
| 7/25 |
Han |
ML Reinforcement Learning (2) |
tutorial |
[/] |
| 7/18 |
Yiqing |
ML Supervised Learning (1) |
tutorial |
/ |
| 7/4 |
Ding |
Bridging the Edge-Cloud Barrier for Real-time Advanced Vision Analytics |
HotCloud19 |
Slides |
| 7/4 |
Weiyan |
ML-tutorial-0 |
Machine Learning |
Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 6/27 |
Xinchen |
Loom: Flexible and Efficient NIC Packet Scheduling |
NSDI19 |
Slides |
| 6/27 |
Hao |
Pantheon: the training ground for Internet congestion-control research
|
Sigcomm18 |
Slides |
| 6/20 |
Ding |
free discuss |
free discuss |
No Slides |
| 6/20 |
Zhuoyi |
free discuss |
free discuss |
No Slides |
| 6/13 |
Yiqing |
ML for CC |
free discuss |
No Slides |
| 6/13 |
Duowen |
free discuss |
free discuss |
No Slides |
| 6/6 |
Jingrong |
free discuss |
free discuss |
No Slides |
| 6/6 |
Qinghe |
free discuss |
free discuss |
No Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 5/30 |
Qinghuan |
free discuss |
free discuss |
No Slides |
| 5/30 |
Tianhan |
free discuss |
free discuss |
No Slides |
| 5/16 |
Weiyan |
free discuss |
free discuss |
No Slides |
| 5/16 |
Chaoliang |
free discuss |
free discuss |
No Slides |
| 5/9 |
Zhaoxiong |
free discuss |
free discuss |
No Slides |
| 5/9 |
Ge |
free discuss |
free discuss |
No Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 4/25 |
Yiding |
free discuss |
free discuss |
No Slides |
| 4/25 |
Hao |
free discuss |
free discuss |
No Slides |
| 4/18 |
Han |
free discuss |
free discuss |
No Slides |
| 4/18 |
Xinchen |
free discuss |
free discuss |
No Slides |
| 4/11 |
Zhuoyi |
free discuss |
free discuss |
No Slides |
| 4/11 |
Ding |
free discuss |
free discuss |
No Slides |
| 4/4 |
Wenxin |
free discuss |
free discuss |
No Slides |
| 4/4 |
Gaoxiong |
free discuss |
free discuss |
No Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 3/28 |
Gaoxiong |
free discuss |
free discuss |
No Slides |
| 3/28 |
Yiqing |
free discuss |
free discuss |
No Slides |
| 3/16 |
Jiacheng |
free discuss |
free discuss |
No Slides |
| 3/16 |
Jingrong |
free discuss |
free discuss |
No Slides |
| 3/9 |
Qinghe |
Secure Federated Transfer Learning |
Topic |
Slides |
| 3/9 |
Duowen |
Tiresias: A GPU Cluster Manager for Distributed Deep Learning |
NSDI19 |
Slides |
| 3/2 |
Jingrong |
Deconstructing RDMA-enabled Distributed Transactions: Hybrid is Better ! |
OSDI18 |
Slides |
| 3/2 |
Ge |
FlowBlaze: Stateful Packet Processing in Hardware |
NSDI19 |
Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 2/23 |
Gaoxiong |
HyperLoop: Group-Based NIC-Offloading to Accelerate Replicated Transactions in Multi-Tenant Storage Systems |
Sigcomm2018 |
Slides |
| 2/23 |
Chaoliang |
LegoOS: A Disseminated, Distributed OS for Hardware Resource Disaggregation |
OSDI18 |
Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 1/19 |
Zhaoxiong |
TicTac: Improving Distributed Deep Learning With Communication Scheduling |
SysML19 |
Slides |
| 1/19 |
Jiacheng |
Selecta: Heterogeneous Cloud Storage Configuration for Data Analytics |
ATC18 |
Slides |
| 1/19 |
Yiqing |
Improving TCP Congestion Control with Machine Intelligence |
SIGCOMM18-NetAI |
Slides |
| 1/12 |
Yiding |
Neural Adaptive Content-aware Internet Video Delivery |
OSDI18 |
Slides |
| 1/12 |
Han |
Mainstream : Dynamic Stem-Sharing for Multi-Tenant Video Processing |
ATC18 |
Slides |
| 1/12 |
Xinchen |
Floem: A Programming System for NIC-Accelerated Network Applications |
OSDI18 |
Slides |
| 1/5 |
Weiyan |
Neural Architecture Search topic |
AAAI |
Slides |
| 1/5 |
Qinghuan |
Applied Machine Learning at Facebook: A Datacenter Infrastructure Perspective |
HPCA |
Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 12/29 |
Weiyan |
AutoML Introduction |
NIPS |
Slides |
| 12/29 |
Ding |
JANUS:Fast and Flexible Deep Learning
via Symbolic Graph Execution of Imperative Programs |
NSDI 2019 |
JANUS,Slides |
| 12/22 |
Wenxin |
Distributed Big Data Stream Processing |
OSDI 2018 |
Three Steps, Slides |
| 12/22 |
Duowen |
Datacenter RPCs can be General and Fast |
CMU & Intel Lab |
eRPC,Slides |
| 12/01 |
Qinghe |
Pretzel: Opening the Black Box of Machine Learning Prediction Serving Systems |
OSDI 2018 |
Pretzel,Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 10/27 |
Yiding |
Device Placement with RL |
ICML 2018 |
Spotlight, Slides |
| 10/27 |
Jiacheng |
TVM |
OSDI 2018 |
TVM, Slides |
| 10/20 |
Han |
Deep Reinforcement learning Overview |
Topic |
Slides |
| 10/13 |
Xinchen |
HotCocoa: Hardware Congestion Control Abstractions |
HotNets 17 |
HotCocoa, Slides |
| 10/06 |
Ding |
Hierarchical Heavy Hitter |
SIGCOMM 2017 |
Heavy Hitter Paper, Slides |
| 10/06 |
Yiqing |
A tutorial on Parameter Server (Part II) |
OSDI 2014 |
Slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 09/29 |
Junxue |
How to use our TestBed |
Tutorial |
TestBed slides |
| 09/22 |
Yiqing |
A tutorial on Parameter Server (Part I) |
OSDI 2014 |
PS paper, PPT |
| 09/08 |
Zhaorong |
Andromeda: Performance, Isolation, and Velocity at Scale in Cloud Network Virtualization |
NSDI 2018 |
Andromeda paper and slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 07/28 |
Junxue |
SnailTrail: Generalizing Critical Paths for Online Analysis of Distributed Dataflows |
NSDI 2018 |
SnailTrail paper and slides |
| 07/21 |
Gaoxiong |
Multi-Path Transport for RDMA in Datacenters |
NSDI 2018 |
MP-RDMA paper and slides |
| 07/14 |
Qinghe |
Chameleon: Scalable Adaptation of video analytics |
Sigcomm 2018 |
Chameleon paper |
| 07/07 |
Jiacheng |
Monotasks: Architecting for Performance Clarity in Data Analytics Frameworks |
SOSP 2017 |
Monotask paper |
| Date |
Presenter |
Topic |
Source |
Links |
| 06/30 |
Justinas |
APUS: Fast and Scalable Paxos on RDMA |
SoCC 2017 |
APUS paper |
| 06/23 |
Shuihai |
Homa: A Receiver-Driven Low-Latency Transport Protocol Using Network Priorities |
Sigcomm 2018 |
HOMA paper |
| 06/16 |
Ge |
KV-Direct: High-Performance In-Memory Key-Value Store with Programmable NIC |
SOSP 2017 |
KV paper slides |
| 06/09 |
Yuchao |
Low Latency RNN Inference with Cellular Batching |
Eurosys 2018 |
batchmaker paper |
| 06/02 |
Wenxin |
Introduction to SmartNICs |
|
slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 05/26 |
Junxue |
NetChain: Scale-Free Sub-RTT Coordination |
NSDI 2018 |
NetChain paper and slides |
| 05/19 |
Hong |
Semi-Oblivious Traffic Engineering: The Road Not Taken |
NSDI 2018 |
SMORE paper and slides |
| 05/12 |
Ahmed |
Exploring Existing Open Source Network Operating Systems and APIs |
|
slides |
| 05/05 |
Chaoliang |
PASTE: A Network Programming Interface for Non-Volatile Main Memory |
NSDI 2018 |
PASTE paper and slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 04/28 |
Jiacheng |
Approximating Fair Queueing on Reconfigurable Switches |
NSDI 2018 |
AFQ paper and slides |
| 04/20 |
Li, Yuchao |
AUTO and BDS rehearsals |
|
|
| 04/14 |
Zhaoxiong |
SilkRoad: Making Stateful Layer-4 Load Balancing Fast and Cheap Using Switching ASICs |
Sigcomm 2017 |
Silkroad paper Silkroad slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 03/24 |
Gaoxiong |
The Case for Moving Congestion Control Out of the Datapath |
HotNets 2017 |
CCP paper |
| 03/17 |
Yiding |
Clipper: A Low-Latency Online Prediction Serving System |
NSDI 2017 |
Clipper paper and slides |
| 03/10 |
Yiqing |
Pingmesh: A Large-Scale System for Data Center Network Latency Measurement and Analysis |
Sigcomm 2015 |
Pingmesh paper |
| 03/03 |
Baochen |
LITE Kernel RDMA Support for Datacenter Applications |
SOSP 2017 |
LITE paper |
| Date |
Presenter |
Topic |
Source |
Links |
| 02/10 |
Qinghe |
NEVE: Nested Virtualization Extensions for ARM |
SOSP 2017 |
NEVE paper slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 01/13 |
Justinas |
Eris: Coordination-Free Consistent Transactions Using In-Network Concurrency Control |
SOSP 2017 |
Eris paper |
| 01/06 |
Li |
Taking the Edge off with Espresso Scale, Reliability and Programmability for Global Internet Peering |
Sigcomm 2017 |
Espresso paper |
| Date |
Presenter |
Topic |
Source |
Links |
| 12/31 |
Ge |
NetCache: Balancing Key-Value Stores with Fast In-Network Caching |
SOSP 2017 |
NetCache paper P4 introduction slides |
| 12/09 |
Junxue |
Carousel: Scalable Traffic Shaping at End Hosts |
Sigcomm 2017 |
Carousel paper |
| 12/02 |
Hong |
RotorNet: A Scalable, Low-complexity, Optical Datacenter Network |
Sigcomm 2017 |
RotorNet paper slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 11/25 |
Bairen |
CrystalNet: Faithfully Emulating Large Production Networks |
SOSP 2017 |
CrystalNet paper slides |
| 11/18 |
Zhaoxiong |
CherryPick: Adaptively Unearthing the Best Cloud Configurations for Big Data Analytics |
NSDI 2017 |
CherryPick paper and slides |
| 11/11 |
Jiacheng |
SLAQ: Quality-Driven Scheduling for Distributed Machine Learning |
SoCC 2017 |
SLAQ paper |
| 11/04 |
Yiding |
No! Not Another Deep Learning Framework |
HotOS 2017 |
paper link |
| Date |
Presenter |
Topic |
Source |
Links |
| 10/28 |
Baochen |
Poseidon: An Efficient Communication Architecture for Distributed Deep Learning on GPU Clusters |
ATC 2017 |
Poseidon paper and slides |
| 10/21 |
Qinghe |
Neural Adaptive Video Streaming with Pensieve |
Sigcomm 2017 |
Pensive paper |
| 10/14 |
Shuihai |
Re-architecting datacenter networks and stacks for low latency and high performance |
Sigcomm 2017 |
NDP paper |
| 10/07 |
Gaoxiong |
The QUIC Transport Protocol: Design and Internet-Scale Deployment |
Sigcomm 2017 |
QUIC paper slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 09/30 |
Li |
Deploying Unique DL Networks as Micro-Services with TensorRT, user Extensible Layers, and GPU Rest Engine |
NVDIA GTC 2017 |
TensorRT slides |
| 09/16 |
Yuchao |
DRILL: Micro Load Balancing for Low-latency Data Center Networks |
Sigcomm 2017 |
DRILL paper slides |
| 09/09 |
Junxue |
Summary of sigcomm 2017 |
Sigcomm 2017 |
papers |
| 09/02 |
Justinas |
vCorfu: A Cloud-Scale Object Store on a Shared Log |
NSDI 2017 |
vCorfu paper and slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 08/26 |
Ge |
Stateless Network Functions: Breaking the Tight Coupling of State and Processing |
NSDI 2017 |
StatelessNF paper and slides |
| 08/19 |
Hong |
resilient datacenter load balancing in the wild |
Sigcomm 2017 |
Hermes paper slides |
| 08/12 |
Gaoxiong |
Evaluating the Power of Flexible Packet Processing for Network Resource Allocation |
NSDI 2017 |
FlexSwitch paper and slides |
| Date |
Presenter |
Topic |
Source |
Links |
| 07/29 |
Bairen |
Performance Isolation Anomalies in RDMA |
KBNets 2017 |
slides |
| 07/15 |
Baochen |
Fast, Scalable and Simple Distributed Transactions with Two-Sided (RDMA) Datagram RPCs |
SOSP 2016 |
Faast paper and slides |