Net2Text: Query-Guided Summarization of Network Forwarding Behaviors
summary
Net2Text converts a natural-language query into an internal SQL-like query over forwarding state and traffic stats, formulates summarization as an NP-hard ILP, and uses a reduced search graph plus sampling to approximate solutions. Implementation uses a rule-based parser and template NLG, supports four query types, and handles hundreds of routers with full routing tables in seconds; with one in 1,000 sampling it returns answers under a second at ~96–97% of the no-sampling score. The system is practical and operator feedback is positive.
AI-written summary · about
chapters (8)
- 0:00Intro
- 1:26Understanding how the network behaves can take hours
- 4:59The database maintains the forwarding state and traffic statistics
- 6:00Each cluster represents a path specification A summary consists of multiple path specifications
- 6:45The translation uses templates to obtain natural language from path specifications
- 12:05Due to the size of the search space exhaustive exploration is not feasible
- 14:21Network traffic is highly skewed across multiple levels
- 20:16THE ADVANCED COMPUTING SYSTEMS ASSOCIATION
related talks
18mNetKeeper: Enhancing Network Resilience with Autonomous Network Configuration...
17mNetHint: White-Box Networking for Multi-Tenant Data Centers
27mJuggling the Jigsaw: Towards Automated Problem Inference from Network Trouble Tickets
30mUsing Python to design, configure, and measure large-scale networks
25mDataplane equivalence and its applications
24mNetHide: Secure and Practical Network Topology Obfuscation
18mActive Learning of Symbolic NetKAT Automata
18mMaking Logic a First-Class Citizen in Generative ML for Networking