Meetings
- Lecture: Monday and Wednesday 1:45-3:00pm, Wilkinson 132
- Instructor: Danyang Zhuo (email)
- TA: Jingrong Chen (email)
Course description and grading
- This is a graduate-level seminar course on data center and cloud systems. In each meeting, we discuss one research paper. There will be 2 students presenting and leading discussion, and you are required to read the paper before the meeting and participate actively in the discussion. Before each meeting, you need to write a paper review.
- You need to complete one research project. Team size is up to 4 students. By the end of the semester, your team need to finish a 6-page research paper in USENIX format. The research project must contain a substantial software development component. You can pick one of the following styles:
- Survey one area in computer systems, and compare performance among candidate software
- Model, improve, or build a mini-version of an existing research paper
- Preliminary evaluation of some early-stage ideas
- Talk to the instructor if you don’t have a project idea.
- Your grade is computed in the following way:
- Warmup programming assignment (10%)
- Paper presentation (10%)
- Participation in discussion (15%)
- Paper review (15%)
- Research Project (50%) = Proposal (5%) + Checkpoint (5%) + Presentation and Report (40%)
Resources
Schedule
Our discussions are centered around three major themes: (1) distributed systems, (2) computer networking, and (3) cluster management.
| Date |
Paper |
Discussion Lead |
Notes |
| Wed, Jan 05 |
MapReduce: Simplified Data Processing on Large Clusters (OSDI 2004) |
Danyang |
Programming assignment released (warmup assignment) |
| Mon, Jan 10 |
Dryad: Distributed Data-Parallel Programs from Sequential Building Blocks (EuroSys 2007) |
Danyang |
|
| Wed, Jan 12 |
Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing (NSDI 2012) |
Danyang |
|
| Mon, Jan 17 |
No Class (MLK day) |
|
|
| Wed, Jan 19 |
GraphX: Graph Processing in a Distributed Dataflow Framework (OSDI 2014) |
Ruiqi, Meihong |
Project group due |
| Mon, Jan 24 |
Scaling Distributed Machine Learning with the Parameter Server (OSDI 2014) |
Sridurga, Charlotte |
|
| Wed, Jan 26 |
TensorFlow: A System for Large-Scale Machine Learning (OSDI 2016) |
Yanming, Minzhi |
Programming assignment due |
| Mon, Jan 31 |
Ray: A Distributed Framework for Emerging AI Applications (OSDI 2018) |
Mingyuan |
|
| Wed, Feb 02 |
Project group meeting to finalize proposal |
|
Project proposal due |
| Mon, Feb 07 |
Arrakis: The Operating System is the Control Plane (OSDI 2014) |
Weicheng, Sasha |
|
| Wed, Feb 09 |
Snap: a Microkernel Approach to Host Networking (SOSP 2019) |
Ziyan, Xiangzhou |
|
| Mon, Feb 14 |
The Design and Implementation of Open vSwitch (NSDI 2015) |
Danyang |
|
| Wed, Feb 16 |
Andromeda: Performance, Isolation, and Velocity at Scale in Cloud Network Virtualization (NSDI 2018) |
Yitu |
|
| Mon, Feb 21 |
FaSST: Fast, Scalable and Simple Distributed Transactions with Two-Sided (RDMA) Datagram RPCs (OSDI 2016) |
Yongxin, Hongwei |
|
| Wed, Feb 23 |
Datacenter RPCs can be General and Fast (NSDI 2019) |
Ke |
|
| Mon, Feb 28 |
Project group meeting to finalize project checkpoint |
|
|
| Wed, Mar 02 |
Project group meeting to finalize project checkpoint |
|
Project checkpoint due |
| Mon, Mar 07 |
Spring Break |
|
|
| Wed, Mar 09 |
Spring Break |
|
|
| Mon, Mar 14 |
ZygOS: Achieving Low Tail Latency for Microsecond-scale Networked Tasks (SOSP 2017) |
Yi, Zhe |
|
| Wed, Mar 16 |
Shenango: Achieving High CPU Efficiency for Latency-sensitive Datacenter Workloads (NSDI 2019) |
Danyang |
|
| Mon, Mar 21 |
Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center (NSDI 2011) |
Yezhou, Shuyan |
|
| Wed, Mar 23 |
Dominant Resource Fairness: Fair Allocation of Multiple Resource Types (NSDI 2011) |
Xunyu, Jingjing |
|
| Mon, Mar 28 |
Large-Scale Cluster Management at Google with Borg (EuroSys 2015) |
Zhu, Ruikang |
|
| Wed, Mar 30 |
Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms (SOSP 2017) |
Aritra, Sohini |
|
| Mon, Apr 04 |
Project group meeting to work on project |
|
|
| Wed, Apr 06 |
Project group meeting to work on project |
|
|
| Mon, Apr 11 |
Project Presentation |
|
|
| Wed, Apr 13 |
Project Presentation |
|
Project report due |