Ad Server Congestion Reduction via SCTE-35 Pre-allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face congestion and dropped requests due to the flooding of ad servers with simultaneous requests for advertising placement in national-based television programming, leading to inefficiencies in advertising decision-making.
Innovation Solution
A computer-implemented method and system that identifies targeted advertisements for recipient signal streams by pre-allocating placement opportunities, using SCTE-35 packets to indicate advertisement placement opportunities, and storing these advertisements in caches, allowing for efficient insertion into signal streams based on audience attributes and channel identifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If ad servers receive simultaneous requests for advertising placement in national-based television programming, then advertising placement decisions can be made, but server congestion and dropped requests occur
Solution Approach 1:
The patent segments the signal stream into multiple sub-signal streams and distributes ad requests across them. This segmentation reduces the load on any single server instance, preventing congestion while maintaining processing capacity. The system divides the monolithic ad serving task into smaller, manageable units that can be handled concurrently by multiple servers.
Solution Approach 2:
The patent implements pre-allocation of placement opportunities and pre-fetching of advertisements before the actual ad insertion point is reached. By performing these actions in advance, the system reduces the computational burden during real-time ad serving, preventing server overload while ensuring ads are ready for immediate insertion.
2Adaptability or versatility
If ad requests are processed in real-time for every signal stream, then accurate targeted advertising can be delivered, but server overload and dropped requests occur
Solution Approach 1:
The patent merges multiple sub-signal streams that share common characteristics (such as the same ad placement opportunities) into consolidated processing groups. This allows the system to process targeted advertising requirements for multiple streams simultaneously, reducing redundant computations while maintaining accurate audience-based targeting for each individual stream.
3Loss of time
If advertisements are fetched and processed at the last moment before insertion, then ad freshness is maintained, but server response time increases and requests are dropped
Solution Approach 1:
The patent implements pre-fetching of advertisements based on predicted ad insertion points and pre-allocation of placement opportunities. By fetching and processing ads in advance rather than at the last moment, the system reduces real-time latency while ensuring ads are ready for immediate insertion, thereby preventing request drops due to server unavailability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer implemented method and system for accelerating advertising placement decisions is disclosed. A server receives a source signal stream having a channel identifier and an indication that an advertisement placement opportunity is located in the signal stream a time period prior to an advertisement signal point. The server identifies a list representative of a plurality of recipient signal streams having the channel identifier and at least one audience attribute prior to expiration of the time period. The server obtains a plurality of targeted advertisements corresponding to the list representative of the plurality of recipient signal streams. The server stores the plurality of advertisements in at least one cache.