DASH Media Segment Key Pre-fetching for Thundering Herd Mitigation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The 'thundering herd problem' occurs when many users simultaneously request decryption keys and initialization vectors (IVs) for media content segments, leading to significant delays and inefficiencies in content delivery, especially during high-demand events like sporting events.

Innovation Solution

Pre-fetching decryption keys and IVs before their availability time and allowing clients to derive IVs using ECB mode encryption, reducing the load on key servers and improving content delivery efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If decryption keys and IVs are requested simultaneously by many users during high-demand events, then all users can access the content, but significant delays occur due to server overload

Engineering Contradiction:

Improvecontent delivery efficiencyVSAvoiddelay in content delivery

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-fetching decryption keys and IVs before they are actually needed for content decryption. Clients request and cache these cryptographic parameters in advance, so when content becomes available, decryption can occur immediately without waiting for key server responses, thereby eliminating playback delays during high-demand events

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If many users simultaneously request decryption keys from the key server, then all users can decrypt content, but the key server becomes overloaded causing thundering herd problem

Engineering Contradiction:

Improvesystem capacity to handle concurrent requestsVSAvoidkey server processing efficiency

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

Clients perform preliminary key acquisition by requesting and caching decryption keys and IVs in advance before peak demand occurs. This distributes the timing of key server requests across multiple time points rather than concentrating them simultaneously, preventing server overload while maintaining the ability to serve many concurrent users

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by allowing clients to autonomously manage their own key caching and decryption processes without requiring real-time key server intervention during playback. Once keys are cached, clients can independently decrypt and play back content without further server interaction, dramatically reducing ongoing key server load

Inventive Principle:

Principle #25Self-service

3Reliability

If decryption keys are obtained at the moment of content request, then security is maintained, but playback latency increases due to additional network round-trip time

Engineering Contradiction:

Improvesecurity of content protectionVSAvoidnetwork round-trip delay

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary key acquisition by requesting and caching decryption keys and IVs in advance of actual content playback needs. This pre-fetching eliminates the need for real-time network round-trips during playback, ensuring both security (keys are still obtained through authorized channels) and low latency (keys are already available when content arrives)

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9900289B2Low-latency secure segment encryption and authentication interface

Publication Date: 2018.02.20 FUTUREWEI TECHNOLOGIES INC

AI summary

An apparatus comprising a memory, a processor coupled to the memory and configured to obtain a Dynamic Adaptive Streaming over Hypertext Transfer Protocol (HTTP) (DASH) Media Presentation Description (MPD) from an HTTP server, wherein the MPD describes a media presentation as at least one encrypted segment, and wherein the encrypted segment is associated with an availability time and a decryption key, and pre-fetch the decryption key associated with the encrypted segment by requesting the decryption key from a key server prior to the availability time of the encrypted segment.