Cold Stream Storage Parameters for Data Retention Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage solutions do not account for the differing access frequencies of hot and cold streams, leading to increased susceptibility of cold streams to data retention, read disturb, and temperature cross issues.
Innovation Solution
Classify data streams as hot or cold based on access frequency, using custom data storage parameters for cold streams and a stream cache partition for hot streams, with eviction policies for the cache partition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the same data storage parameters are used for both hot streams and cold streams, then the data storage device maintains simple operation and general-purpose storage, but cold streams become more susceptible to data retention issues, read disturb issues, and temperature cross issues
Solution Approach 1:
The patent applies local quality by implementing different data storage parameters specifically for cold streams while maintaining general parameters for hot streams. The controller detects cold streams based on access frequency and applies specialized storage parameters (such as different program voltages, pulse widths, or write patterns) only to those streams, thereby improving data retention reliability for cold streams without requiring complete system-wide parameter changes.
Solution Approach 2:
The patent implements parameter changes by dynamically adjusting data storage parameters based on stream classification. The controller modifies storage parameters (e.g., programming voltage levels, write speed, verification thresholds) according to the detected access frequency of each stream, transforming the storage behavior adaptively to match the specific needs of cold versus hot streams.
2Reliability
If cold stream data is rewritten frequently to improve resilience to storage issues, then data reliability improves, but device lifetime decreases due to increased program erase cycles
Solution Approach 1:
The patent applies partial action by performing selective rewriting of cold stream data rather than frequent rewriting of all data. The controller identifies cold streams through access frequency detection and applies specialized storage parameters and targeted rewriting operations only to those specific streams, avoiding unnecessary program erase cycles on hot streams and other data, thus balancing reliability improvement with device lifetime preservation.
3Speed
If a stream cache partition is implemented for hot streams with eviction policies, then hot stream access performance improves, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the storage device into distinct functional partitions: a stream cache partition for hot streams and a primary storage partition for cold streams and other data. The controller implements separate management mechanisms for each partition, including eviction policies for the cache partition and specialized storage parameters for the primary partition, thereby improving hot stream access performance while organizing complexity into manageable segments.
Solution Approach 2:
The patent implements dynamics by creating a dynamic stream cache partition that adapts to changing access patterns. The controller continuously monitors access frequency and dynamically adjusts which streams reside in the cache partition versus the primary storage partition, applying eviction policies that respond to real-time usage patterns, thereby optimizing access speed while managing complexity through adaptive behavior.
Data Source
AI summary
A stream classification system of a data storage device monitors an access frequency of data stored by the data storage device. If the data is classified as cold data, the stream classification system rewrites the cold data to a primary storage partition using one or more special data storage parameters. The one or more special data storage parameters cause the cold data to be more resilient to various data storage issues including, but not limited to, data retention issues, read disturb issues and/or temperature cross issues.