Dynamic I/O Routing Between Storage Pools to Reduce Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern storage systems experience increased I/O latency due to throttling limits, which can lead to queue buildup and delayed processing, as primary storage pools become overwhelmed, while alternate pools with available capacity are underutilized for read operations.
Innovation Solution
A method is implemented where an I/O manager determines when a primary storage pool is throttled and identifies an alternate pool with available processing capacity, altering the I/O command flow to send commands to the alternate pool for processing, thereby reducing overall latency by utilizing underutilized resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If I/O commands are processed only by the primary storage pool, then data consistency is maintained, but I/O latency increases due to throttling limits
Solution Approach 1:
The patent introduces an I/O manager as an intermediary component that sits between the host and storage pools. This manager dynamically routes I/O commands to either the primary or alternate storage pool based on real-time processing capacity assessment, thereby mediating between the need for data consistency (primary pool) and the need for low latency (alternate pool when primary is throttled).
Solution Approach 2:
The system dynamically adjusts the routing of I/O commands based on the real-time processing capacity of storage pools. The I/O manager continuously monitors throttling status and processing rates, adapting the command routing strategy accordingly - sending commands to the primary pool when it has capacity, and to the alternate pool when the primary is throttled, thus optimizing both consistency and latency performance.
2Reliability
If I/O commands are queued during throttling, then data integrity is preserved, but processing speed decreases due to queue buildup
Solution Approach 1:
The patent segments the storage system into multiple independent storage pools (primary and alternate), each capable of processing I/O commands independently. This segmentation allows the system to bypass queue buildup in the primary pool by routing commands to the alternate pool, thereby maintaining data integrity while improving processing speed during throttling conditions.
Solution Approach 2:
The I/O manager acts as an intermediary that assesses the processing capacity of storage pools and dynamically routes commands accordingly. When the primary pool is throttled and queues are building up, the I/O manager redirects commands to the alternate pool, preventing queue buildup delays while ensuring data integrity through coordinated access.
3Reliability
If alternate storage pools are used for backup only, then data safety is ensured, but resource utilization decreases due to underutilization
Solution Approach 1:
The patent makes the alternate storage pool multi-functional by enabling it to serve both its traditional backup function and as an active I/O processing pool when the primary is throttled. This universality allows the alternate pool to utilize its idle capacity for processing read commands, thereby improving resource utilization while maintaining data safety through its continued role in replication and backup.
Solution Approach 2:
The system enables the alternate storage pool to serve itself by processing I/O commands directly when capacity is available, rather than always relying on the primary pool. This self-service capability allows the alternate pool to utilize its idle resources for productive work, improving overall resource utilization while maintaining its backup function.
Data Source
AI summary
A computer-implemented method to reduce Input/Output (I/O) command latency. The method includes receiving, by a first storage pool, a plurality of I/O commands, wherein the first storage pool includes a first throttling limit, and the first throttling limit. The method further includes determining a first processing rate of the first storage pool is at the first throttling limit. The method also includes determining a second processing rate of a second storage pool is below a second throttling limit, wherein the second storage pool is communicatively coupled to the first storage pool. The method includes sending, by the first storage pool to the second storage pool, one or more of the plurality of I/O commands. The method further includes processing, by the second storage pool, the one or more I/O commands of the plurality of I/O commands. The method includes returning processed I/O commands to the host.