Index

Datacenter Networks

Architecture

• A scalable, commodity, data center network architecture - Al-Fares et al., SIGCOMM '08

RDMA

• Anuj Kalia's PhD thesis provides a good overview of emerging network hardware/technology

• Using One-Sided RDMA Reads to Build a Fast, CPU-Efficient Key-Value Store - Mitchell et al., ATC '13

• FaRM: Fast Remote Memory - Dragojević et al., NSDI '14

• Design Guidelines for High Performance RDMA Systems - Kalia et al., ATC '16

• Revisiting Network Support for RDMA - Mittal et al., SIGCOMM '18

• Datacenter RPCs can be General and Fast - Kalia et al., NSDI '19

• FreeFlow: Software-based Virtual RDMA Networking for Containerized Clouds - Kim et al., NSDI '19

Kernel

• netmap: a novel framework for fast packet I/O - Rizzo ATC '12

• IX: A protected dataplane operating system for high throughput and low latency - Belay et al., OSDI '14

• Arrakis: the operating system is the control plane - Peter et al., OSDI '14

• mTCP: a Highly Scalable User-level TCP Stack for Multicore Systems - Jeong et al., NSDI' 14

• Understanding Host Network Stack Overheads - Cai et al., SIGCOMM '21

Programmable Networks

• IncBricks: Toward In-Network Computation with an In-Network Cache - Liu et al., ASPLOS '17

• NetCache: Balancing key-value stores with fast in-network caching - Jin et al., SOSP '17

• NetChain: Scale-Free Sub-RTT Coordination - Jin et al., NSDI '18

• Offloading Distributed Applications onto SmartNICs using iPipe - Ming et al., SIGCOMM '19

• E3: Energy-Efficient Microservices on SmartNIC-Accelerated Servers - Ming et al., ATC '19

• When Should The Network Be The Computer? - Ports et al., HotOS' 19

  • Discussed what the in-network computing should be used for

• Pegasus: Tolerating Skewed Workloads in Distributed Storage with In-Network Coherence Directories - OSDI' 20

• Gallium: Automated Software Middlebox Offloading to Programmable Switches - Zhang et al., SIGCOMM '20

  • Automattically partitions an input software middlebox into a P4 program that runs on a programmable switch and an x86 non-offloaded program that runs on a regular server

• TEA: Enabling State-Intensive Network Functions on Programmable Switches - Kim et al., SIGCOMM '20

  • Allows NFs on programmable switches to look up large virtual address built on external DRAM.

  • Key ideas: RDMA + bounded linear probing

• Scaling Distributed Machine Learning with In-Network Aggregation - Sapio et al., NSDI '21

• ATP: In-network Aggregation for Multi-tenant Learning - Lao et al., NSDI '21

Wide Area Networks

Video Streaming

• Improving Fairness, Efficiency, and Stability in HTTP-based Adaptive Video Streaming with FESTIVE - Jiang et al., CoNEXT '12

• A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service - Huang et al., SIGCOMM '14

• Neural Adaptive Video Streaming with Pensieve - Mao et al., SIGCOMM '17

• Salsify: Low-Latency Network Video through Tighter Integration between a Video Codec and a Transport Protocol - Fouladi et al., NSDI '18

  • Proposes a tightly coupled codec and transport protocol

  • Exploits its codec's ability to save and restore its internal state

  • Three options when sending the next frame(lower quality frame/higher quality frame/skip)

  • Never send a frame unless the network is ready

• Neural Adaptive Content-aware Internet Video Delivery - Yeo et al., OSDI '18

• Vantage: optimizing video upload for time-shifted viewing of social live streams - Ray et al., SIGCOMM '19

  • Retransmit low-quality frames during high bandwidth period to improve QoE for delayed viewers

• Learning in situ: a randomized experiment in video streaming - Yan et al., NSDI' 20

• Neural-Enhanced Live Streaming: Improving Live Video Ingest via Online Learning - Kim et al., SIGCOMM '20

Misc

• The Design Philosophy of the DARPA Internet Protocols - Clark SIGCOMM '88

• The Road to SDN: An Intellectual History of Programmable Networks - Feamster et al., CCR '14