Distributed Storage I/O Limit Control for Faster Data Copy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed storage systems face challenges in efficiently managing data copy operations without exceeding node performance limits, which can impact the production environment's performance.
Innovation Solution
A management system that calculates and sets upper I/O performance limits for data copy processes based on node performance information, load monitoring, and volume arrangements to optimize data copy speed while minimizing interference with production tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If data copy operations are performed at high speed, then data copy time is reduced, but node performance limits are exceeded causing degradation of production system performance
Solution Approach 1:
The management system proactively determines and sets upper I/O performance limits for data copy processes before actual data copy operations begin. By calculating available I/O performance based on node performance information and current load conditions, and setting appropriate upper limits in advance, the system prevents performance degradation of production systems before it occurs, rather than reacting after the problem manifests
Solution Approach 2:
The system dynamically adjusts the upper I/O performance limit parameter based on changing system conditions. By monitoring current node loads and recalculating available I/O performance, the management system modifies the upper limit parameter to match current capacity, enabling faster data copy when resources are available while maintaining production system stability when resources are constrained
2Productivity
If data copy operations are performed at high speed, then productivity of data copy process is improved, but harmful effects on production environment increase
Solution Approach 1:
The management system continuously monitors current node loads and uses this feedback to dynamically adjust upper I/O performance limits. By comparing actual system state against available I/O performance calculations, the system modifies data copy speed limits in real-time, maximizing data copy productivity when resources permit while automatically reducing impact on production environments when resources are constrained
Solution Approach 2:
The system transitions from static I/O performance limits to dynamic, adaptive limits that respond to changing system conditions. By continuously recalculating available I/O performance based on current node loads and adjusting upper limits accordingly, the system optimizes data copy speed dynamically, achieving high productivity during low-utilization periods while preventing harmful impacts during high-utilization periods
Data Source
AI summary
A management system shortens the length of time required for data copy while reducing its effects on other processes. The management system manages a storage system including a plurality of nodes. The management system references data copy information, application information, and volume information to identify a target node offering a target volume for a data copy process. The management system determines an available input/output performance value of the target node on the basis of node performance information and monitoring information. The management system determines an upper input/output performance limit of the data copy process on the basis of the available input/output performance value of the target node.