IO Workload Prioritization in Data Mirroring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data mirroring technologies face challenges in efficiently selecting and prioritizing IO workloads in networked storage arrays, leading to suboptimal performance and throughput in data replication and disaster recovery scenarios.
Innovation Solution
A method for IO workload selection and prioritization is introduced, where weights are set and adjusted for different classes of IO operations (SYNC, ASYNC, and ADP) to optimize resource allocation, ensuring that IOs are processed based on predetermined ratios, with a focus on ordered workflow processing to maximize network bandwidth and minimize latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If randomized interleaved workflow IO selection is used, then device complexity is reduced, but IO response time and network throughput deteriorate
Solution Approach 1:
The patent segments IO workloads into three distinct classes (SYNC, ASYNC, ADP) with different priority levels and processing requirements. By dividing the IO processing into separate queues and applying class-specific weights, the system achieves ordered workflow processing that improves response time without requiring complex randomized selection mechanisms.
Solution Approach 2:
The patent dynamically adjusts processing weights for different IO classes based on system state and performance requirements. By changing the parameter weights assigned to SYNC, ASYNC, and ADP IOs, the system optimizes throughput and response time without needing complex selection algorithms, achieving better performance with simpler mechanics.
2Loss of time
If weights are set for ordered workflow processing, then IO response time improves, but device complexity increases
Solution Approach 1:
The patent applies different processing qualities and weights to different IO classes locally. SYNC IOs receive higher priority weights for critical responses, while ASYNC and ADP IOs receive lower weights. This localized quality differentiation improves overall response time without requiring global complex coordination mechanisms.
Solution Approach 2:
The patent implements dynamic weight adjustment for different IO classes based on system conditions and performance targets. The weights for SYNC, ASYNC, and ADP IOs can be modified to optimize throughput and response time, providing adaptability without complex structural changes to the processing mechanism.
3Productivity
If multiple IO classes are processed simultaneously without prioritization, then productivity is maintained, but network throughput deteriorates
Solution Approach 1:
The patent segments network traffic into distinct IO classes (SYNC, ASYNC, ADP) that are processed in an ordered workflow. This segmentation allows the system to maintain high productivity by processing multiple IO types while optimizing network throughput through prioritized handling of critical SYNC operations followed by ASYNC and ADP operations.
Solution Approach 2:
The patent uses weight parameters to control the processing mix of different IO classes. By adjusting these weights, the system can optimize the balance between productivity (total IO processing volume) and network throughput (efficient data replication), achieving both goals simultaneously through parameter optimization rather than trade-offs.
Data Source
AI summary
A method of IO workload selection and prioritization in data mirroring environments includes setting a first weight, a second weight, and a third weight to be used during an iteration of IO processing. The method further includes processing IOs of a first class of IO operations while there are IOs of the first class of IO operations to be processed and while a first weight for the first class of IO operations is positive, processing IOs of a second class of IO operations while there are IOs of the second class of IO operations to be processed and while a second weight for the second class of IO operations is positive, processing IOs of a third class of IO operations while there are IOs of the third class of IO operations to be processed and while a third weight for the third class of IO operations is positive.