Dynamic Socket Segmentation for Data Replication Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed storage systems face challenges in efficiently managing data replication messages, particularly in achieving low latency for synchronous data replication due to the use of multiple types of network sockets with different queue depths, which affects overall system performance.
Innovation Solution
The system transitions between states, initially using both type-1 and type-2 network sockets for data replication and then switches to using only type-2 sockets when a synchronous data replication message is detected, effectively reducing the number of sockets used for transmission and minimizing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple types of network sockets with different queue depths are used for data replication, then the system can handle both synchronous and asynchronous replication messages, but the transmission latency for synchronous messages increases
Solution Approach 1:
The patent segments the socket handling by creating separate socket instances for synchronous and asynchronous data replication. Type-1 sockets are dedicated to synchronous replication with lower queue depth, while type-2 sockets handle asynchronous replication with higher queue depth. This segmentation eliminates the latency penalty of using high queue depth sockets for synchronous operations, as each segment uses appropriately optimized socket characteristics for its specific replication type.
2Loss of time
If type-1 sockets with lower queue depth are used for synchronous replication, then transmission latency is reduced, but the system cannot efficiently handle asynchronous replication messages
Solution Approach 1:
The system segments socket resources into type-1 sockets optimized for synchronous replication (lower queue depth for reduced latency) and type-2 sockets optimized for asynchronous replication (higher queue depth for improved throughput). This allows each replication type to utilize sockets with appropriate characteristics, achieving both low latency for synchronous operations and high efficiency for asynchronous operations simultaneously.
Solution Approach 2:
The patent implements dynamic socket selection based on the replication type of incoming messages. The system dynamically determines whether to use type-1 or type-2 sockets depending on whether the message requires synchronous or asynchronous replication, optimizing performance characteristics adaptively rather than using a fixed socket configuration for all replication operations.
Data Source
AI summary
A method is disclosed comprising: instantiating a first plurality of type-1 network sockets on a source computing system, each of the type-1 network sockets having a first queue depth; instantiating a second plurality of type-2 network sockets on the source computing system, each of the type-2 network sockets having a second queue depth that is greater than the first queue depth; transitioning the source computing system into a first state, the first state being one in which data replication messages are transmitted by the source computing system to a target computing system by using both the type-1 and type-2 network sockets; transitioning the source computing system from the first state into a second state, the second state being one in which data replication messages are transmitted by the source computing system to the target computing system by using the type-2 network sockets.