Deduplication Using Low-Latency Random Read Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional deduplication methods on disk devices lead to increased read latency due to the need for the read/write head to frequently seek and retrieve blocks from different tracks, as they store redundant and non-redundant blocks on different tracks, resulting in significant seek times.
Innovation Solution
The use of low-latency random read memory (LLRRM) devices, such as flash memory, to store and retrieve matching blocks, reducing the need for seek operations and thereby mitigating read latency by allowing random read operations without track-based seek times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If deduplication is implemented on disk devices by storing redundant and non-redundant blocks on different tracks, then storage space is saved through redundancy removal, but read latency increases due to frequent seek operations between different tracks
Solution Approach 1:
The patent segments the storage system into two distinct components: disk devices for storing non-redundant blocks and LLRRM devices for storing redundant blocks. This segmentation allows each component to be optimized for its specific function, with the LLRRM handling random read requests for redundant blocks without causing seek operations on the disk device.
Solution Approach 2:
The LLRRM acts as an intermediary between the disk device and the read request for redundant blocks. Instead of directly accessing redundant blocks on the disk device which would require seek operations, the system uses LLRRM as a mediator to serve these requests, thereby eliminating seek time for redundant block reads.
2Productivity
If the read/write head frequently seeks to retrieve blocks from different tracks for deduplication reads, then all blocks can be retrieved from the disk device, but significant seek times are incurred
Solution Approach 1:
The LLRRM serves as an intermediary that handles random read requests for redundant blocks, eliminating the need for the read/write head to seek between different tracks on the disk device. This mediator approach allows efficient data retrieval without incurring seek times.
Solution Approach 2:
The patent replaces the mechanical seek operation of the read/write head with an electronic/random access mechanism provided by the LLRRM. Instead of physically moving the read/write head to different tracks, the system uses the random read capability of LLRRM to retrieve redundant blocks, substituting mechanical movement with electronic access.
3Quantity of substance
If matching blocks are stored on disk devices for deduplication, then storage space is utilized, but random read operations experience high latency due to track-based access requirements
Solution Approach 1:
The patent segments the storage function by assigning different roles to different storage devices: disk devices store non-redundant blocks while LLRRM devices store redundant blocks. This segmentation enables optimized access patterns where random reads for redundant blocks are served by LLRRM's fast random access capability rather than the slower disk device.
Solution Approach 2:
The patent changes the access parameter by introducing LLRRM with different access characteristics (low-latency random read) to complement the disk device. This parameter change allows the system to achieve both high storage capacity and fast random read speeds for redundant blocks.
Data Source
AI summary
Deduplication of data using a low-latency random read memory (LLRRM) is described herein. Upon receiving a block, if a matching block stored on a disk device is found, the received block is deduplicated by producing an index to the address location of the matching block. In some embodiments, a matching block having a predetermined threshold number of associated indexes that reference the matching block is transferred to LLRRM, the threshold number being one or greater. Associated indexes may be modified to reflect the new address location in LLRRM. Deduplication may be performed using a mapping mechanism containing mappings of deduplicated blocks to matching blocks, the mappings being used for performing read requests. Deduplication described herein may reduce read latency as LLRRM has relatively low latency in performing random read requests relative to disk devices.