Peer-Based Video Server Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current load-balancing techniques introduce a single point of failure and are inefficient in managing bandwidth-intensive, time-sensitive systems like video-on-demand, as they rely on simple parameters and lack proactive server status updates, leading to unpredictable load distribution and potential server overload.
Innovation Solution
A peer-based load-balancing system where each server maintains state information about other servers in its group, proactively updates this information, and uses advanced algorithms to redirect requests based on multiple parameters such as content availability, memory, and cache status, eliminating the single point of failure and optimizing load distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a supervisor/subordinate load-balancing architecture is employed, then request distribution is simplified, but a single point of failure is introduced reducing system reliability
Solution Approach 1:
The patent divides the load-balancing function into multiple independent peer servers rather than having a single supervisor. Each server maintains state information about others and can independently handle requests, eliminating the single point of failure while distributing the coordination function across multiple nodes.
Solution Approach 2:
The patent combines the load-balancing and state-maintenance functions into each peer server itself, rather than separating these functions from the data servers. Each server becomes both a data server and a load-balancing participant, eliminating the need for a separate supervisor while maintaining coordination capabilities.
2Ease of manufacture
If DNS round robin load balancing is used, then implementation is simple, but load distribution becomes unpredictable and servers may become overloaded
Solution Approach 1:
The patent implements proactive state information exchange where servers notify each other of their current status (load, content availability, cache status). This feedback mechanism allows servers to make informed load-balancing decisions based on real-time conditions rather than simple round-robin sequencing, improving load distribution efficiency while maintaining peer-to-peer simplicity.
3Reliability
If secondary supervisors are added to handle failures, then system reliability improves, but device complexity and administrative overhead increase
Solution Approach 1:
The patent enables each peer server to autonomously determine its own status and make load-balancing decisions without requiring a supervisor. Servers self-manage their state information and self-coordinate with peers, eliminating the need for primary/secondary supervisor architecture while maintaining high availability through distributed decision-making.
4Speed
If simple load parameters are used for distribution, then processing is fast, but bandwidth-intensive time-sensitive content delivery is inefficient
Solution Approach 1:
The patent expands the state information parameters beyond simple load counts to include content availability, cache status, and bandwidth capacity. This allows servers to make more informed decisions about bandwidth-intensive content delivery, directing requests to servers with appropriate resources while maintaining fast processing through pre-exchanged state information.
Data Source
AI summary
A system and method for load balancing a plurality of servers is disclosed. In a preferred embodiment, a plurality of servers in a video-on-demand or other multi-server system are divided into one or more load-balancing groups. Each server preferably maintains state information concerning other servers in its load-balancing group including information concerning content maintained and served by each server in the group. Changes in a server's content status or other state information are preferably proactively delivered to other servers in the group. When a content request is received by any server in a load-balancing group, it evaluates the request in accordance with a specified algorithm to determine whether it should deliver the requested content itself or redirect the request to another server in its group. In a preferred embodiment, this determination is a function of information in the server's state table.