https://www.usenix.org/conference/nsdi22/technical-sessions
Cluster resource management
Efficient Scheduling Policies for Microsecond-Scale Tasks
A Case for Task Sampling based on Learning for Cluster Job Scheduling
Starlight: Fast Container Provisioning on the Edge and over the WAN
Transport layer
PowerTCP: Pushing the Performance Limits for Datacenter Networks
RDMA is Turing complete, we just did not know it yet!
FlexTOE: Flexible TCP Offload with Fine-Grained Parallelism
An edge-queued datagram service for all datacenter traffic
Backpressure Flow Control
Packet Order Matters! Improving Application Performance by Deliberately Delaying Packets
Video streaming
Swift: Adaptive Video Streaming with Layered Neural Codecs
Ekya: Continuous Learning of Video Analytics Models on Edge Computer Servers
Yuzu: Neural-Enhanced Volumetric Video Streaming
Programmable switches
NetVRM: Virtual Register Memory for Programmable Netowrks
SwiSh: Distributed Shared State Abstractions for Programmable Switches
Modular Switch Programming Under Resource Constraints
Enabling In-situ Programmability in Network Data Plane: From Architecture to Language
Runtime Programmable Swtiches
IMap: Fast and Scalable In-Network Scanning with Programmable Switches
Unlocking the Power of Inline Floating-Point Operations on Programmable Switches
Security and Privacy
Privid: Practical, Privacy-Preserving Video Analytics Queries
Spectrum: High-bandidth Anonymous Broadcast
Donar: Anonymous VoIP over Tor
Network troubleshooting and debugging
Closed-loop Network Performance Monitoring and Diagnosis with SpiderMon
Collie: Finding Performance Anomalies in RDMA subsystems
SCALE: Automatically Finding RFC Compliance Bugs in DNS Nameservers
Operational track
Decentralized cloud wide-area network traffic engineering with BLASTSHIELD
Detecting Ephemeral Optical Events with OpTel
Bluebird: High-performance SDN for Bare-metal Cloud Services
Cetus: Releasing P4 Programmers from the Chore of Trial and Error Compiling
Wireless
Exploiting Digital Micro-Mirror Devices for Ambient Light Communication
Whisper: IoT in the TV White Space Spectrum
Learning to Communicate Effectively Between Battery-free Devices
Saiyan: Design and Implementation of a Low-power Demodulator for LoRa Backscatter Systems
Reliable distributed systems
Graham: Synchronizing Clocks by Leveraging Local Clock Properties
IA-CCF: Individual Accountability for Permissioned Ledgers
DispersedLedger: High-Throughput Byzantine Consensus on Variable Bandwidth Networks
Raising the bar for programmable hardware
Re-architecting Traffic Analysis with Neural Network Interface Cards
Elixir: A High-performance and Low-cost Approach to Managing Hardware/Siftware Hybrid Flow Tables Considering Flow Burstiness
Gearbox: A Hierarchical Packet Scheduler for Approximate Weighted Fair Queueing
Testing and verification
Performance Interfaces for Network Functions
Automated Verification of Network Function Binaries
Differential Network Analysis
Katra: Realtime Verificiation for Multilayer Networks
Sketch-based telemetry
Cloud scale services
Accelerating Collective Communication in Data Parallel Training across Deep Learning Frameworks
Cocktail: A Multidimensional Optimization for Model Serving in Cloud
Data-Parallel Actors: A Programming Model for Scalable Query Serving Systems
Orca: Server-assisted Multicast for Datacenter Networks
ISPs and CDNs
Yeti: Stateless and Generalized Multicast Forwarding
cISP: A Speed-of-Light Internet Service Provider
Configanator: A Data-driven Approach to Improving CDN Performance
C2DN: How to Harness Erasure Codes at the Edge for Efficient Content Delivery
Cloud Scale Resource Management
Optimizing Netowrk Provisioning through Cooperation
OrbWeaver: Using IDLE Cycles in Programmable Networks for Opportunistic Coordination
CloudCluster: Unearthing the Functional Structure of a Cloud Service
Data center network infrastructure
Zeta: A Scalable and Robust East-West Communication Framework in Large-scale Clouds
Aquila: A unified, low-latency fabric for datacenter networks
RDC: Energy-Efficient Data Center Network COngestion Relief with Topological Reconfigurability at the Edge
Multitenancy
Isolation Mechanisms for High-Speed Packet-Processing Pipelines
Justitia: Software Multi-Tenancy in Hardware Kernel-Bypass Networks
NetHint: White-Box Networking for Multi-Tenant Data Centers
Software switching and beyond
Tiara: A Scalable and Efficient Hardware Acceleration Architecture for Stateful Layer-4 Load Balancing
Scaling Open vSwitch with a Computational Cache
Backdraft: a Lossless Virtual Switch that Prevents the Slow Receiver Problem
Last updated 4 years ago