Variable-Block Data Archive for Storage Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data storage systems using fixed block architectures face inefficiencies in space utilization and capacity due to fixed block sizes, leading to wasted capacity and performance limitations.
Innovation Solution
Implementing a data archive architecture with remote head servers that manage file storage metadata in a database, allowing data to be written and read sequentially across multiple storage devices, enabling higher utilization of storage space and scalability by partitioning data into portions and distributing them across multiple devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed block architecture is used for data storage, then data can be stored in uniform blocks, but storage space utilization becomes inefficient and capacity is reduced
Solution Approach 1:
The patent divides data into variable-size blocks instead of fixed blocks, allowing each block to be sized according to the actual data requirements. This segmentation approach eliminates the wasted space that occurs with fixed-block architectures, thereby improving storage capacity while maintaining organized data structure.
Solution Approach 2:
The system changes the block size parameter from fixed to variable, allowing block sizes to adapt to different data types and requirements. This parameter flexibility enables more efficient space utilization and increases overall storage capacity without sacrificing the ease of data organization.
2Quantity of substance
If data is partitioned and distributed across multiple storage devices, then storage capacity and scalability increase, but system complexity increases
Solution Approach 1:
The patent introduces a metadata server as an intermediary that manages the distribution and tracking of data blocks across multiple storage devices. This mediator handles the complexity of data partitioning and location tracking, allowing the system to scale across multiple devices without proportionally increasing overall system complexity.
Solution Approach 2:
The metadata server performs multiple functions including block allocation, location tracking, and data reconstruction coordination. This multi-functional approach consolidates management tasks into a single component, reducing the complexity that would otherwise be distributed across multiple specialized systems.
3Speed
If shingled magnetic recording is used to prioritize data retrieval, then read performance improves, but write capability is reduced
Solution Approach 1:
The system performs preliminary actions by pre-positioning data blocks in optimal locations and pre-establishing retrieval pathways during the write phase. This allows the SMR drives to operate in their optimized read mode while maintaining acceptable write performance through advance preparation of data layouts.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for massive scalability, efficient use of storage space, and improved performance by prioritizing data retrieval over writing, enabling the storage of large amounts of data across hundreds of exabytes while maintaining data integrity and redundancy.
Implementation Method 1
The storage devices can utilize shingled magnetic recording (SMR) to compactly write data to a disk track in a manner that writes to both the data-destination track and also the next adjacent track
Data Source
AI summary
Systems and methods for reading data are provided herein using a data archive architecture controlled by a head server. The head server can read data in a contiguous manner across multiple storage devices for data that has been partitioned into multiple portions and spread across the multiple storage devices. Portions of a first object can be transmitted and read from a first subset of storage devices and portions of a second object can be transmitted and read from a second subset of storage devices. The head server can increment a read pointer based on a length of the portions of the first object to a determine a read location of different portions of the first object or the second object. The head server can identify different read location to read sub sequent of different portions of multiple objects using one or more lengths of the objects or portions of the object.