Green Storage Tiering for Cold Data Energy Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems fail to effectively reduce power consumption based on data access patterns, leading to unnecessary energy usage.
Innovation Solution
Implement a 'green tier' in data storage systems where rarely accessed 'cold' data is moved to dedicated devices that can be powered off or placed in a low-power state, leveraging existing power management functions to conserve energy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all production storage devices operate in normal power mode to ensure data accessibility, then data access reliability is improved, but power consumption increases
Solution Approach 1:
The storage system is segmented into two distinct tiers: production storage devices that remain in normal power mode for active data, and green tier devices that operate in low-power mode for cold data. This segmentation allows the system to maintain reliability for frequently accessed data while reducing power consumption for rarely accessed data.
Solution Approach 2:
The system dynamically transitions storage devices between normal and low-power modes based on data access patterns. The green tier devices can be dynamically activated when cold data becomes hot, ensuring reliability is maintained when needed while enabling power savings during low-activity periods.
2Measurement precision
If production storage devices are continuously monitored and scanned to identify cold data, then data classification accuracy is improved, but system complexity increases
Solution Approach 1:
The storage system performs self-service by automatically monitoring its own usage patterns and identifying cold data without requiring external intervention. The system autonomously classifies data based on access patterns and manages the tiering process, reducing the need for complex external management infrastructure.
Solution Approach 2:
The system implements feedback loops where usage patterns are continuously monitored and fed back into the tiering decision process. This feedback mechanism enables accurate identification of cold data while automating the classification process, reducing manual complexity.
3Use of energy by moving object
If cold data is moved to green tier devices operating in low-power mode, then power consumption is reduced, but data access speed may deteriorate
Solution Approach 1:
Different storage tiers are assigned different quality characteristics appropriate to their function: production devices provide high-speed access for active data, while green tier devices provide energy-efficient storage for cold data. Each tier is optimized for its specific purpose, ensuring that speed requirements are met where necessary.
Solution Approach 2:
The system performs preliminary actions by proactively identifying and moving data to appropriate tiers before access is needed. Cold data is pre-positioned on green tier devices, and the system maintains metadata about data locations and access patterns to enable quick retrieval even from low-power devices.
Data Source
AI summary
A data storage system is operated for reduced power consumption, including regularly scanning usage patterns of data stored on a set of production storage devices to identify cold data whose usage is below a predetermined usage threshold, and adding the cold data of each scan to a collection of cold data. The production storage devices are operated in a normal power mode having normal per-device power consumption. Upon the collection of cold data reaching a predetermined size, (1) one of the production storage devices is designated as a green tier device, and (2) the collection of cold data is moved to the green tier device. The green tier device is operated in a low power mode having a lower per-device power consumption.