SAN Server Load Distribution via Distance Pairing Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the television broadcast industry, Storage Area Networks (SANs) face limitations in supporting high data rates for multiple broadcast quality video streams due to the number of fibre channel log-ins, leading to the need for load balancing and redundancy, but existing solutions either result in server failures or increased complexity and cost.
Innovation Solution
A system that distributes the client load among remaining servers using a distance pairing distribution algorithm, ensuring that the failure of one server increases the load on other servers by no more than a minimal amount, with dual client network connections and redundant network interfaces, maintaining redundancy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If direct attached storage with fibre channel connections is used to support high data rates, then data transfer speed is improved, but the number of supported servers is limited due to fibre channel log-in limitations
Solution Approach 1:
The system segments the storage network into direct-attached storage servers and network-attached storage clients. Servers that require high-speed access maintain direct fibre channel connections to storage devices, while other servers access storage through the Ethernet-based storage area network. This segmentation allows the system to support both high data transfer rates for critical servers and a larger overall server population through the network.
2Productivity
If load balancing is implemented across Ethernet connections, then server utilization is improved, but redundant connections require additional complexity and configuration
Solution Approach 1:
The system introduces a storage area network with virtualization layer as an intermediary between servers and physical storage devices. This intermediary abstracts the complexity of direct storage connections, allowing servers to access storage resources through standardized network protocols without requiring complex fibre channel configurations. The virtualization layer handles load balancing and resource allocation, simplifying the overall system architecture.
3Reliability
If redundant servers are configured for failover, then system reliability is improved, but the cost and complexity of the system increases
Solution Approach 1:
The system creates universal access to storage resources through the storage area network, where any server can access any storage device through the network infrastructure. This multi-functionality eliminates the need for dedicated redundant physical connections for each server, as the network itself provides failover capabilities. Storage resources can be dynamically allocated and accessed by any server in the network, reducing overall system complexity while maintaining high availability.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A system delivers data and includes a plurality of data storage devices. A plurality of servers such as video servers are connected to the plurality of data storage devices and access the data storage devices to provide streaming data to a plurality of clients upon request therefrom. Upon failure of one of the servers, the load on the remaining servers is distributed equally among remaining servers using a distribution algorithm that assigns dual client network connections to server groupings such that the failure of any one server will increase the client load on any remaining server by no more than ceil (C/P) loads, where C is the total number of configured clients that access the plurality of servers in the system and P is the total number of server pairs.