Concurrent Data Replication Scheduling via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face challenges in efficiently scheduling and balancing large replication tasks with other service level objectives, as traditional replication methods can be time-consuming and resource-intensive, and do not effectively adjust priorities as task requirements change.
Innovation Solution
A distributed system that divides replication tasks into multiple portions, schedules each portion using a global scheduling queue, and dynamically adjusts priorities to ensure efficient processing and fairness among tasks, allowing for concurrent task completion and adherence to service level objectives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional replication methods are used, then data replication can be completed, but the process is time-consuming and resource-intensive
Solution Approach 1:
The patent divides a large replication task into multiple smaller portions or segments. Each segment can be processed independently and concurrently, allowing the system to replicate data more efficiently by utilizing multiple processing channels simultaneously rather than handling the entire dataset as a single monolithic task.
Solution Approach 2:
The patent implements dynamic priority adjustment where the system can change the priority of replication tasks based on changing requirements and conditions. This allows the system to adapt to varying service level objectives and resource availability, optimizing replication efficiency by dynamically reassigning processing resources to high-priority segments.
2Productivity
If traditional replication methods are used, then data replication can be completed, but resource utilization is not optimized
Solution Approach 1:
By segmenting replication tasks into smaller portions, the system can distribute work across multiple processing units and optimize resource allocation. This allows for better utilization of available computational resources, preventing any single resource from being overloaded while ensuring continuous productive work across the system.
Solution Approach 2:
The patent changes the parameter of task granularity by dividing large replication tasks into smaller segments. This parameter change enables the system to adjust resource allocation dynamically, improving overall resource utilization efficiency while reducing the computational burden on individual processing units.
3Adaptability or versatility
If traditional replication methods are used, then replication tasks can be executed, but priorities cannot be adjusted as task requirements change
Solution Approach 1:
The patent implements a dynamic scheduling system where task priorities can be adjusted in real-time based on changing service level objectives and requirements. This dynamic capability allows the system to adapt to varying conditions without requiring complete reconfiguration, balancing adaptability with manageable system complexity through automated priority management.
Solution Approach 2:
The patent incorporates feedback mechanisms that monitor task progress and system conditions, automatically adjusting priorities based on this feedback. This allows the system to adapt to changing requirements intelligently, reducing the need for manual intervention while maintaining appropriate task prioritization despite increased scheduling complexity.
4Productivity
If large replication tasks are processed as single units, then task management is simple, but concurrent processing and fairness among tasks cannot be achieved
Solution Approach 1:
The patent segments large replication tasks into smaller portions that can be processed concurrently. This segmentation enables multiple tasks to execute in parallel, improving overall system productivity and fairness among competing tasks, while the segmentation itself provides a natural structure for managing scheduling complexity.
Data Source
AI summary
A system, computer program product, and computer-executable method of scheduling replication on a data storage system, the system, the computer program product, and computer-executable method including receiving a request to replicate a portion of data stored on the data storage system, dividing the portion of data into a plurality of portions, scheduling replication of each of the plurality of portions, and replicating each of the plurality of portions.