~/talks

Ravi Netravali

what they talk about

all 12 in the filters

where they keep going back

  • NSDI2 editions2020–2023

all 7 in the filters

on a stage with

all 64 in the filters

who talks about the same things

nobody close enough yet

titleconferencedate
Mowgli: Passively Learned Rate Control for Real-Time VideoNSDI '252025-04-28
Mowgli trains a rate controller entirely offline from production GCC telemetry, using soft actor critic plus conservative Q-learning and distributional critics…
Talk on
MadEye: Boosting Live Video Analytics Accuracy with Adaptive Camera Configurations21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24)2024-04-16
MadEye is a camera–server system that actively reorients commodity PTZ cameras, runs a fast search over rotations/zooms, and uses on‑camera lightweight…
Talk on
Sprinter: Speeding Up High-Fidelity Crawling of the Modern Web21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24)2024-04-16
Sprinter combines a small browser-crawled subset with a browserless phase that replays results from a computation cache built by rewriting JS and recording…
Talk on
ADR-X: ANN-Assisted Wireless Link Rate Adaptation for Compute-Constrained Embedded...21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24)2024-04-16
System achieves 10× lower packet loss (down to 0.1–0.5% versus typical 2–10%) while running ~100× faster than prior RL work, with inference taking 170–380 μs…
Talk on
NetVigil: Robust and Low-Cost Anomaly Detection for East-West Data Center Security21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24)2024-04-16
NetVigil uses flow-level logs aggregated into IP-granularity communication graphs (300 vertices vs ~139k flow-vertex graphs) and a GNN autoencoder trained with…
Talk on
Canvas: Isolated and Adaptive Swapping for Multi-Applications on Remote MemoryNSDI 20232023-04-17
Canvas isolates per-application swap data paths and builds a replacement swap system on Linux 5.5 that gives each workload dedicated swap partition, swap…
Talk on
Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNsNSDI 20232023-04-17
Bamboo makes preemptible (spot) instances usable for large DNN training by building a distributed system that inserts redundant computations into idle pipeline…
Talk on
RECL: Responsive Resource-Efficient Continuous Learning for Video AnalyticsNSDI 20232023-04-17
System integrates a shared model zoo of previously adapted lightweight experts, a learned model selector that predicts per-frame accuracy and aggregates…
Talk on
Boggart: Towards General-Purpose Acceleration of Retrospective Video AnalyticsNSDI 20232023-04-17
Boggart builds a preprocessor that uses classical CV—background subtraction, morphological ops, connected components, keypoints and descriptors—to produce…
Talk on
Gemel: Model Merging for Memory-Efficient, Real-Time Video Analytics at the EdgeNSDI 20232023-04-17
Gemel is a system that jointly retrains and merges layers and weights across edge vision models so one copy of shared layers fits in GPU memory, combined with…
Talk on
Dashlet: Taming Swipe Uncertainty for Robust Short Video StreamingNSDI 20232023-04-17
Implements out-of-order chunk pre-buffering that schedules specific chunks of upcoming short videos based on a simple non-ML swipe model derived from two…
Talk on
Privid: Practical, Privacy-Preserving Video Analytics Queries19th USENIX Symposium on Networked Systems Design and Implementation (NSDI '22)2022-04-04
Privid defines ρ,K,ε-event-duration privacy and enforces it by splitting video into chunks, running analyst-provided black-box neural nets per chunk, then…
Talk on
Horcrux: Automatic JavaScript Parallelism for Resource-Efficient Web Computation15th USENIX Symposium on Operating Systems Design and Implementation (OSDI '21)2021-07-14
Horcrux automatically parallelizes legacy page JavaScript by doing offline concolic + dynamic analysis on servers to produce conservative per-function…
Talk on
Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and15th USENIX Symposium on Operating Systems Design and Implementation (OSDI '21)2021-07-14
Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads John Thorpe, Yifan Qiao, Jonathan Eyolfson, and…
Talk on
Alohamora: Reviving HTTP/2 Push and Preload by Adapting Policies On the FlyNSDI ’212021-04-12
Alohomora uses reinforcement learning to pick per-load HTTP/2 push and preload actions, trains one model per cluster of structurally similar pages, and…
Talk on
ABC: A Simple Explicit Congestion Controller for Wireless NetworksNSDI 20202020-02-25
ABC: A Simple Explicit Congestion Controller for Wireless Networks Prateesh Goyal, MIT CSAIL; Anup Agarwal, CMU; Ravi Netravali, UCLA; Mohammad Alizadeh and…
Talk on
Fawkes: Faster Mobile Page Loads via App Inspired Static TemplatingNSDI 20202020-02-25
Fawkes extracts static templates from multiple HTML versions offline, serves those templates immediately from cache, and streams small dynamic patches that…
Talk on
Vesper: Measuring Time-to-Interactivity for Web PagesNSDI '182018-04-09
Proposes a concrete interactivity metric called Ready Index and a rewiring tool Vesper that discovers which DOM nodes and JS heap variables must be loaded for…
Talk on
Prophecy: Accelerating Mobile Page Loads Using Final-state Write LogsNSDI '182018-04-09
Prophecy has the server run a headless browser (PhantomJS-style), rewrite JavaScript with proxy-based shims to record a single final write log entry per JS…
Talk on
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.