Proposal for a Systems Reading Group

We propose a reading group on the purpose of advanced systems education. In this reading group, group members read and present important papers in the field of computer systems.

Rule

• This event will be held once a week.
• During each event, two teams (with two people in each team) will present.
• Each team will present one of the system papers listed below. They will have twenty minutes for the presentation. During the 20-minute presentation, both team members should talk for longer than 5 minutes.
• After the presentation, there will be a 20-minute Q&A and 20-minute group discussion (with up to 6 people in each group). During the Q&A, everyone is encouraged to ask questions and give comments. If noone volunteers, the host of the event will decide who to talk next.
• Before each event, the presenters should upload their slides to this website. The slides should be open-cource under a CC BY-NC-SA 4.0 licenses. The presenters are expected to create the slide themselves; however, they may use any data, text, figure, or code in the original paper.
• Before each event, members of the reading group should study the materials listed below, including watching videos and reading papers.
• Because it is hard to discuss with hundreds of people, this event has a prerequest. To join the group, one need to finish at least 60% of labs of an operating systems course where they are required to implement a whole hardware-bootable operating system from scratch (such as MIT 6.S081).
• The reading group only has 40 slots. If more than 40 people are qualified, they should be splitted into multiple groups.

Paper List and Schedule

W0: Icebreaking

• Introduce yourself, find a teammate, and have fun.

W1: Operating Systems

• The Multikernel: A new OS architecture for scalable multicore systems
• LegoOS: A Disseminated, Distributed OS for Hardware Resource Disaggregation (Is this paper too new? Do we need a more classic one?)

W2: Virtualization

• Live Migration of Virtual Machines
• Memory Resource Management in VMware ESX Server

W3: Distributed Systems

• The Google File System
• MapReduce: Simplified Data Processing on Large Clusters

W4: Program Analysis and Debugging

• KLEE: Unassisted and Automatic Generation of High-Coverage Tests for Complex Systems Programs (Is this the best paper of symbolic execution?)
• ReVirt: enabling intrusion analysis through virtual-machine logging and replay

W5: Security and Pravacy

• Meltdown and Specture (two papers in a pack)
• TaintDroid: an information-flow tracking system for realtime privacy monitoring on smartphones

W6: FPGA and GPU

• What’s the best paper for FPGA? AmorphOS, Coyote, Cascade, or Catapult?
• What’s the best paper for GPU/CUDA?

W7: Networking

• Infiniswap?
• What’s the best networking paper to read? Find one from Facebook?

W8: Machine Learning

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• TVM: An Automated End-to-End Optimizing Compiler for Deep Learning

W9: New Memory

• Transactional memory
• NVM

W10: Formal Verification

• What’s the best and most understandable paper for formal verification?