The Science

Our teams are comprised of tenured experts with a focus on precision and academic integrity. We have written the papers that define the streaming industry and our technology is exact, our data is exceptional, and our innovation is proven.  Understanding how to leverage the internet as a mass-scale video transport medium has positioned Conviva to be the architect of great streaming experiences.

The Patents


Monitoring the performance of a content player
Augmenting the functionality of a content player
Monitoring the performance of a content player
Custom video metrics management platform
Custom traffic tagging on the control plane backend
Distributing information over a network
Dynamic generation of video manifest files
Remediation of the impact of detected synchronized data requests in a content delivery network


Dynamic bitrate range selection in the cloud for optimized video streaming
Virtual resource locator
Centrally coordinated peer assignment
Streaming decision in the cloud
Network insights
Switching content
Switching content
Reassigning source peers


Fault isolation in over-the-top content (OTT) broadband networks
Custom traffic tagging on the control plane backed
Custom video metrics management platform
Pre-viewer engagement-based video optimization
Correlating playback information of video segments
Network insights
Virtual resource locator
Dynamic bitrate range selection in the cloud for optimized video streaming
Facilitating client decisions
Source assignment based on network partitioning


Streaming decision in the cloud
Dynamic client logging and reporting
Advertising engine
Automatic diagnostics alerts


Advanced Resource Selection
Remote multi-target client monitoring for streaming content
Automatic diagnostics alerts for streaming content encoded by multiple entities
Correlating playback information of video segments
Fast Direct Resource Allocation Identifier

The Academic Papers

Time-State Analytics

Our recent academic paper, presented at CIDR 2023, posits that while our Timeline method of measuring streaming video has been incredibly successful the Timeline abstraction method is generally applicable across many more domains and enables new opportunities for further research.

Raising the Level of Abstraction for Time-State Analytics

Video AI Architecture

Our most recent academic research has been centered around artificial intelligence–specifically, artificial intelligence models for better understanding video on the Internet.

Understanding the Impact of Video Quality on User Engagement – Winner ACMSIGCOMM’s 2022 Test of Time Award

Redesigning CDN-broker interactions for improved content delivery

Pytheas: Enabling Data-Driven Quality of Experience Optimization Using Group-Based Exploration-Exploitation

Oboe: Auto-tuning video ABR algorithms to network conditions

A Case for a Coordinated Internet-Scale Video Control Plane

Developing a Predictive Model of Quality of Experience for Internet Video

C3: Internet-Scale Control Plane for Video Quality Optimization

CFA: A Practical Prediction System for Video QoE Optimization


Software-Defined Networks and Intelligent Network Control

Dr. Zhang and Dr. Stoica have worked on new network architectures that enable direct control for routing data traffic, supporting network-level objectives, and providing network-wide visibility.

A Case for End System Multicast

Tesseract: A 4D Network Control Plane

A Clean Slate 4D Approach to Network Control and Management

The Feasibility of Supporting Large-Scale Live Streaming Applications with Dynamic Application End-Points

Managing Data Transfers in Computer Clusters with Orchestra

Big Data and Cloud Computing

Dr. Stoica’s research group at the University of California, Berkeley has done pioneering research on building large-scale, real-time cluster computing frameworks, and on enabling multiple frameworks such as Hadoop and Hypertable, to seamlessly share the same cluster.

A Common Substrate for Cluster Computing

Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing

Improving MapReduce Performance in Heterogeneous Environments

Shark: SQL and Rich Analytics at Scale


Scalable Management and Delivery Systems

Dr. Zhang’s End System Multicast (ESM) research group at Carnegie Mellon pioneered the overlay multicast architecture and techniques to support scalable internet video streaming. The award-winning Ph.D. dissertation of Dr. Ion Stoica, co-founder and chief technical officer, pioneered the first scalable architecture to manage quality of service states in a distributed networking environment.

Chord: A Scalable Peer-to-Peer Lookup Protocol for Internet Applications

Stateless Core: A Scalable Approach for Quality of Service in the Internet

Early Experience with an Internet Broadcast System Based on Overlay Multicast

The Impact of DHT Routing Geometry on Resilience and Proximity


Viewer Experience Optimization

Our award-winning teams of engineers and scientists have discovered and validated the need for a preemptive viewer experience optimization tool that quantifies and validates video quality impact on viewer engagement.

Impact of Delivery Ecosystem Variability and Diversity on Internet Video Quality

Service Disciplines for Packet-Switching Integrated-Services Networks

Improving Fairness, Efficiency and Stability in HTTP-Based Adaptive Video Streaming with Festive

The Awards

50 leaders of the pack streaming media
Blog_CSI_Award_2019-300x150 copy
1 - Content Innovation Awards 2017
3 - TV Connect Shortlisted 2017
8 - VideoNet-Connected-TV-Award 2016
13-TVB-Europe-Best-of-IBC-2012 (1)