Storage System Predicting Backup Periods via Bandwidth Fluctuations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to predict the backup data amount and bandwidth fluctuations in networks with fluctuating bandwidth, making it difficult to determine if remote backups meet the recovery point objective (RPO) requirements, especially in medium- and small-scale companies using inexpensive WANs like ADSL.
Innovation Solution
A storage system that includes a processor and memory to measure, record, and predict usable bandwidth and data amounts, allowing for the calculation of the time required for data copying to a remote location, considering bandwidth fluctuations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If inexpensive WAN such as ADSL is used for remote backup, then cost is reduced, but bandwidth quality deteriorates and fluctuates making RPO prediction difficult
Solution Approach 1:
The system performs preliminary measurement and prediction of backup data amounts and WAN bandwidth before executing backups. By analyzing historical data trends and predicting future values, the system can determine in advance whether RPO requirements will be met, allowing administrators to make informed decisions about backup execution timing and resource allocation.
Solution Approach 2:
The system continuously measures actual backup data amounts and WAN bandwidth during backup operations, compares these with predicted values, and uses this feedback to refine future predictions. This closed-loop approach improves prediction accuracy over time and enables dynamic adjustment of backup strategies to ensure RPO compliance despite bandwidth fluctuations.
2Measurement precision
If bandwidth prediction technology is added to predict backup data amount and bandwidth, then RPO requirement determination becomes possible, but system complexity increases
Solution Approach 1:
The storage system performs self-measurement of backup data amounts and self-monitoring of WAN bandwidth using its own resources. By leveraging existing system components for measurement and prediction tasks, the system avoids the need for separate external measurement devices or complex third-party integration, thereby reducing overall system complexity while maintaining prediction accuracy.
Data Source
AI summary
There is provided a storage system providing a storage volume and being coupled via a network to a secondary storage system for storing a backup of data stored in the storage volume, the storage system copies data stored in the storage volume to the secondary storage system, upon receiving an instruction to create a backup; records the size of the data copied; predict the size of data to be copied in the future based on the recorded size of the copied data; records a usable bandwidth of the network at a time when the data is copied to the secondary storage system; predicts a usable bandwidth in the future based on the recorded usable bandwidth of the network; and predicts a time period required for copying data to the secondary storage system based on the predicted size of data to be copied and the predicted usable bandwidth of the network.