benchmarking#
64 picked talks tagged benchmarking, newest conference first.
Open vSwitch and OVN 2026 Fall Conference#
- Death by Kernel locks: Hyperscalers road to DPDK
- From User Defined Networks to EVPN: Scaling OVN-Kubernetes
- Rebuilding an OpenFlow Controller in Rust
- Route scope delimitation as a control plane scalability mechanism for OVN Interconnect using route tags
@Scale: Product 2026#
- Adopting AV1 for Real-Time Communication (RTC) at Scale
- Beyond “Good Enough”: How deployments accelerate research at the Audio AI Frontier
- Chasing a Moving Target: Going AI-Native on Edits' Video Seek Engine
- Flywheel of Media Models & Product Velocity
- ML Video Codec (MLVC): Research to Product
NANOG 98#
PyTorch Conference 2026#
- Five Bottlenecks, Five Fixes: Systematic Performance Diagnosis for PyTorch Distributed Training
- Scaling Foundation Models: From Broken to Near-Linear Multi-Node Training
OCP Global Summit 2026#
OPI Summit on DPUs/IPUs 2026#
AI Engineer New York 2026#
- Everything will be Simulated
- Frontier Models for Acting in Space and Time
- Not All GPU's Are Deployed The Same
- Your GPU Neighbor Can Hear Your Model
AusNOG 2026#
Netdev 0x1A (Rome, 13-16 July 2026)#
AI Engineer World's Fair 2026#
- Are LLM Performance Benchmarks Reliable?
- Beyond Static Intelligence: Evaluating Continual Learning
- Can LLMs write fast multi-GPU kernels? We built a benchmark to find out.
- Computer Use at the Edge of the Statistical Precipice
- Harnessing Collective Agent Intelligence for Open Science
- Kubernetes Is Not Your Sandbox
- Modular: Taming the AI Hardware Cambrian Explosion
- Seeing the Plumbing: Profiling vLLM Speculative Decoding on NVIDIA Blackwell
@Scale: AI & Data 2026#
- Agentic Inference Optimization for RecSys ML models
- Evaluating Opsmate: Grading AI Incident Investigations at Scale
TNC26 (GÉANT research networking conference)#
- Instant Data Structures via Shared Memory
- Network Flow Monitoring: An Evaluation of Modern Collector Software
NANOG 97#
MLSys 2026#
- MLCommons Chakra: Advancing Performance Benchmarking and Co-design using Standardized Execution Traces
- Optimizing Deployment Configurations for LLM Inference
- SONAR: Benchmarking Topology and Collaboration in Decentralized Learning
NSDI '26#
KubeCon + CloudNativeCon Europe 2026#
GTC 2026#
Cisco Live Europe 2026 (Amsterdam)#
ACM CoNEXT 2025#
- Demystifying Performance of eBPF Network Applications
- MoonEm — High-Precision Path Property Emulation using DPDK
- SepaNet: Towards Practical Large-Scale Full-System Simulation for Systems Research
AutoCon 4 (Austin)#
OVS + OVN 2025 Fall Conference (Prague)#
- AI-Powered Performance Insights: Integrating OVS/OVN automatic performance regression analysis with LLMs
- OpenFlow Classifier: Arcane knowledge and common pitfalls
- OvS and Socket Maps - Beyond ports as endpoints
- Revisiting checksum offloads in OVS
- Turbocharge Kubernetes with offloaded OVN-Kubernetes and OVS on DPUs
KubeCon + CloudNativeCon North America 2025#
DENOG 17 (Essen)#
- "What can possibly go wrong?" Stories of peculiarities of modern server hardware
- "What can possibly go wrong?" Stories of peculiarities of modern server hardware
- Th3 Flock of Bird s
- Th3 Flock of Bird s
RIPE 91 (Bucharest)#
OCP Global Summit 2025#
SIGCOMM 2025#
ITNOG9#
MLSys 2025#
- Context Parallelism for Scalable Million-Token Inference
- Radius: Range-based Gradient Sparsity for Large Foundation Model Pre-training