Selective Data Redundancy for Storage Read Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Read latency in storage devices, such as solid state drives, is increased due to die-contention where reads to a target die are blocked by ongoing operations like write or erase operations, and existing solutions like duplicating all data introduce significant overhead.
Innovation Solution
Storing a redundant copy of data across multiple locations on separate dies or channels, with the storage controller categorizing data as hot or cold to determine whether to duplicate it, allowing read access from a secondary location when the primary is busy, thereby reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If data is duplicated to a second location to reduce read latency, then read latency is reduced, but storage overhead increases
Solution Approach 1:
The patent applies local quality by differentiating between hot data and cold data, applying redundancy only to hot data that requires frequent access. The storage controller categorizes data based on access patterns and applies selective duplication, making the redundancy strategy localized to specific data types rather than universal, thereby reducing overall storage overhead while maintaining performance for critical data.
Solution Approach 2:
The patent implements partial action by duplicating only a portion of the data (specifically, only hot data) rather than all data. The storage controller selectively copies data to a second location based on access frequency and latency requirements, performing redundancy only where necessary to reduce read latency, thus avoiding the excessive overhead of full data duplication.
2Reliability
If data is duplicated across multiple dies, then read access reliability is improved, but device complexity increases
Solution Approach 1:
The patent uses copying by creating redundant copies of hot data in a second location (second die or channel). The storage controller identifies hot data and copies it to an alternative physical location, enabling failover or alternative access paths when the primary die is unavailable, thereby improving read access reliability without requiring complete system redundancy.
Solution Approach 2:
The storage controller acts as an intermediary that manages the complexity of multi-die access. It monitors die availability, determines which data should be accessed from which location, and coordinates read operations across multiple dies. This intermediary layer abstracts the complexity from the host system while enabling reliable access through redundant paths.
3Productivity
If the storage controller monitors and categorizes all data, then data access optimization is improved, but processing overhead increases
Solution Approach 1:
The storage controller performs preliminary action by categorizing data as hot or cold during the write operation or shortly thereafter. By identifying and marking hot data early, the system prepares redundancy information in advance, so that when read requests arrive, the controller can quickly determine which data requires alternative access paths without performing complex analysis at read time, thus optimizing data access while controlling processing overhead.
Data Source
AI summary
Systems and methods for reducing read latency by storing a redundant copy of data are described. In one embodiment, the systems and methods include identifying data assigned to be written to a page of a storage device, storing the data in a page of a first die of the storage device, and saving at least one codeword from the data to a page of a second die. In some embodiments, the first die is associated with a first channel of the storage device and the second die is associated with a second channel of the storage device.