Distributed Storage Power Reduction via Hierarchy Rules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed data storage systems face high power consumption due to redundancy requirements, leading to increased costs, heat generation, and earlier hardware failures, with existing improvements being incremental and focused on hardware rather than data coding and placement strategies.
Innovation Solution
Implementing a hierarchy rule based on a spreading policy and set of tolerable failures to distribute erasure-encoded data across a distributed storage system, allowing selective disabling of portions without affecting data reading or writing, thereby reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If additional storage resources are provided for data redundancy, then data reliability is improved, but power consumption increases
Solution Approach 1:
The storage system is segmented into multiple hierarchical levels (data centers, racks, nodes, devices) with data distributed across these segments. Erasure-coded data chunks are placed strategically across different segments so that a specified number of segments can be powered down while maintaining data accessibility from remaining segments, thus reducing power consumption while preserving reliability.
Solution Approach 2:
The system dynamically manages power consumption by allowing selective powering down of storage portions based on operational conditions. The hierarchy rule enables dynamic determination of which data portions can be disabled, transforming the static redundancy model into a dynamic power management system that adjusts resource allocation based on actual needs.
2Reliability
If storage resources are increased for redundancy, then data protection is improved, but heat generation increases
Solution Approach 1:
By segmenting the storage system into hierarchical levels and distributing erasure-coded data chunks across these segments, the system can isolate and power down specific segments that are not currently needed for data access. This reduces the number of active storage devices, thereby reducing overall heat generation while maintaining data protection through the remaining active segments.
3Reliability
If more storage resources are deployed for failure tolerance, then system reliability is improved, but device lifespan decreases
Solution Approach 1:
The system implements dynamic power management where storage portions are selectively powered down when not needed for data operations. This reduces the cumulative operational time and stress on storage devices, extending their lifespan. The hierarchy rule ensures that sufficient redundancy remains active to maintain system reliability, balancing device usage with reliability requirements.
4Reliability
If data is distributed across multiple portions for redundancy, then fault tolerance is improved, but system complexity increases
Solution Approach 1:
The system segments data into erasure-coded chunks and distributes them across a hierarchical structure of storage portions. The hierarchy rule provides a systematic framework for organizing these segments at different levels (data centers, racks, nodes, devices), making the complex distribution manageable through structured rules that define how many chunks can be disabled at each hierarchical level while maintaining fault tolerance.
Data Source
AI summary
Techniques are described for reducing power consumption in a distributed data storage system using a hierarchy rule that is generated based on a spreading policy and a set of tolerable failures. A method may operate to distribute erasure-encoded data of a first data object across first and second portions of a distributed storage system using a hierarchy rule corresponding to a spreading policy based on a set of tolerable failures from which the first data object can be recovered. The method disables the first portion of the distributed storage system that includes a first portion of the erasure-encoded data. The first portion of the distributed storage system is determined according to the spreading policy and the hierarchy rule identifies the set of tolerable failures.