NSDI '25: 22nd USENIX Symposium on Networked Systems...

NSDI '25: 22nd USENIX Symposium on Networked Systems Design and Implementation

9:10am EDT

PRED: Performance-oriented Random Early Detection for Consistently Stable Performance in Datacenters Liberty Ballroom

9:30am EDT

Rajomon: Decentralized and Coordinated Overload Control for Latency-Sensitive Microservices Liberty Ballroom

9:50am EDT

Learnings from Deploying Network QoS Alignment to Application Priorities for Storage Services Liberty Ballroom

10:10am EDT

DISC: Backpressure Mitigation In Multi-tier Applications With Distributed Shared Connection Liberty Ballroom

11:00am EDT

Quicksand: Harnessing Stranded Datacenter Resources with Granular Computing Liberty Ballroom

11:20am EDT

Beehive: A Scalable Disaggregated Memory Runtime Exploiting Asynchrony of Multithreaded Programs Liberty Ballroom

11:40am EDT

Making Serverless Pay-For-Use a Reality with Leopard Liberty Ballroom

12:00pm EDT

GRANNY: Granular Management of Compute-Intensive Applications in the Cloud Liberty Ballroom

2:00pm EDT

MeshTest: End-to-End Testing for Service Mesh Traffic Management Liberty Ballroom

2:20pm EDT

Preventing Network Bottlenecks: Accelerating Datacenter Services with Hotspot-Aware Placement for Compute and Storage Liberty Ballroom

2:40pm EDT

Enhancing Network Failure Mitigation with Performance-Aware Ranking Liberty Ballroom

3:00pm EDT

One-Size-Fits-None: Understanding and Enhancing Slow-Fault Tolerance in Modern Distributed Systems Liberty Ballroom

3:50pm EDT

Accelerating Design Space Exploration for LLM Training Systems with Multi-experiment Parallel Simulation Liberty Ballroom

4:10pm EDT

Optimizing RLHF Training for Large Language Models with Stage Fusion Liberty Ballroom

4:30pm EDT

Minder: Faulty Machine Detection for Large-scale Distributed Model Training Liberty Ballroom

4:50pm EDT

Holmes: Localizing Irregularities in LLM Training with Mega-scale GPU Clusters Liberty Ballroom

5:10pm EDT

SimAI: Unifying Architecture Design and Performance Tuning for Large-Scale Large Language Model Training with Scalability and Precision Liberty Ballroom

5:30pm EDT

ByteCheckpoint: A Unified Checkpointing System for Large Foundation Model Development Liberty Ballroom

9:00am EDT

AutoCCL: Automated Collective Communication Tuning for Accelerating Distributed and Parallel DNN Training Liberty Ballroom

9:20am EDT

OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud Liberty Ballroom

9:40am EDT

Efficient Direct-Connect Topologies for Collective Communications Liberty Ballroom

10:00am EDT

SuperServe: Fine-Grained Inference Serving for Unpredictable Workloads Liberty Ballroom

10:50am EDT

Learning Production-Optimized Congestion Control Selection for Alibaba Cloud CDN Liberty Ballroom

11:10am EDT

GPU-Disaggregated Serving for Deep Learning Recommendation Models at Scale Liberty Ballroom

11:30am EDT

Evolution of Aegis: Fault Diagnosis for AI Model Training Service in Production Liberty Ballroom

11:50am EDT

PAPAYA Federated Analytics Stack: Engineering Privacy, Scalability and Practicality Liberty Ballroom

2:00pm EDT

ONCache: A Cache-Based Low-Overhead Container Overlay Network Liberty Ballroom

2:20pm EDT

GREEN: Carbon-efficient Resource Scheduling for Machine Learning Clusters Liberty Ballroom

2:40pm EDT

The Benefits and Limitations of User Interrupts for Preemptive Userspace Scheduling Liberty Ballroom

3:00pm EDT

Securing Public Cloud Networks with Efficient Role-based Micro-Segmentation Liberty Ballroom

3:50pm EDT

Understanding and Profiling NVMe-over-TCP Using ntprof Liberty Ballroom

4:10pm EDT

Building an Elastic Block Storage over EBOFs Using Shadow Views Liberty Ballroom

4:30pm EDT

Pushing the Limits of In-Network Caching for Key-Value Stores Liberty Ballroom

9:00am EDT

Verifying maximum link loads in a changing world Liberty Ballroom

9:20am EDT

A Layered Formal Methods Approach to Answering Queue-related Queries Liberty Ballroom

9:40am EDT

Runtime Protocol Refinement Checking for Distributed Protocol Implementations Liberty Ballroom

10:00am EDT

CEGS: Configuration Example Generalizing Synthesizer Liberty Ballroom

10:50am EDT

Suppressing BGP Zombies with Route Status Transparency Liberty Ballroom

11:10am EDT

ValidaTor: Domain Validation over Tor Liberty Ballroom

11:30am EDT

From Address Blocks to Authorized Prefixes: Redesigning RPKI ROV with a Hierarchical Hashing Scheme for Fast and Memory-Efficient Validation Liberty Ballroom

11:50am EDT

PreAcher: Secure and Practical Password Pre-Authentication by Content Delivery Networks Liberty Ballroom

2:00pm EDT

ClubHeap: A High-Speed and Scalable Priority Queue for Programmable Packet Scheduling Liberty Ballroom

2:20pm EDT

Self-Clocked Round-Robin Packet Scheduling Liberty Ballroom

2:40pm EDT

Everything Matters in Programmable Packet Scheduling Liberty Ballroom

3:00pm EDT

When P4 Meets Run-to-completion Architecture Liberty Ballroom

3:50pm EDT

Mutant: Learning Congestion Control from Existing Protocols via Online Reinforcement Learning Liberty Ballroom

4:10pm EDT

CATO: End-to-End Optimization of ML-Based Traffic Analysis Pipelines Liberty Ballroom

4:30pm EDT

Resolving Packets from Counters: Enabling Multi-scale Network Traffic Super Resolution via Composable Large Traffic Model Liberty Ballroom

4:50pm EDT

BFTBrain: Adaptive BFT Consensus with Reinforcement Learning Liberty Ballroom