Priority-Aware Internode Messaging for Active-Active Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data storage systems with active-active configurations, the increased throughput from solid-state drives has made internode messaging a significant consumer of processing and communication resources, becoming a limiting factor in performance.
Innovation Solution
A technique where a first storage node delays sending non-latency-critical messages to a second storage node until a latency-critical message is ready, combining these messages into a single aggregated message for transmission, reducing the overall number of messages and associated overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If storage nodes send messages frequently to maintain synchronization in active-active configurations, then synchronization reliability is improved, but processing and communication overhead increases
Solution Approach 1:
The patent combines multiple non-latency-critical messages into a single aggregated message that is transmitted together with a latency-critical message. This merging approach reduces the total number of message transmissions while ensuring that critical synchronization messages are delivered promptly, thereby maintaining reliability without excessive overhead.
Solution Approach 2:
The patent implements a message queue mechanism where non-latency-critical messages are held in queue temporarily before transmission. This preliminary action allows the system to batch these messages and send them together with critical messages, reducing communication overhead while ensuring critical messages are not delayed.
2Stability of the object's composition
If the number of messages between storage nodes is increased to maintain synchronization, then data consistency is improved, but communication resource consumption increases
Solution Approach 1:
Multiple messages are merged into a single aggregated message transmission. Non-latency-critical messages are combined with latency-critical messages in the same transmission packet, reducing the number of separate communications required while maintaining data consistency through proper message ordering and acknowledgment mechanisms.
3Loss of time
If messages are sent immediately without aggregation, then message delivery timeliness is improved, but message overhead increases
Solution Approach 1:
The system performs preliminary classification of messages into latency-critical and non-latency-critical categories. Critical messages are queued for immediate transmission, while non-critical messages are held in a separate queue for later aggregation, reducing overall message overhead without delaying critical communications.
Solution Approach 2:
The message aggregation mechanism is dynamic rather than static. The system adaptively determines when to aggregate messages based on the presence of latency-critical messages, adjusting the aggregation behavior in real-time to balance timeliness and overhead requirements.
Data Source
AI summary
A technique for managing messaging between storage nodes of a storage system includes a first storage node delaying the sending of non-latency-critical messages to a second storage node until the first storage node has a latency-critical message to be sent. The technique further includes combining the non-latency-critical messages with the latency-critical message to form a single, aggregated message, and sending the aggregated message to the second storage node.