Storage Core Pre-Caching Using Learned Request Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage devices face inefficiencies in managing finite bandwidth and resources for pre-caching data for accelerator cores, leading to data thrashing and resource wastage due to unpredictable data request patterns.
Innovation Solution
A storage device with a storage core that learns request patterns and analyzes usage of previously cached data by accelerator cores, optimizing pre-cache operations to efficiently manage resources and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the controller pre-caches data based on access patterns, then data access latency is reduced, but data thrashing occurs causing resource and energy waste
Solution Approach 1:
The patent implements feedback mechanisms where the controller monitors data access patterns from accelerator cores and uses this information to dynamically adjust pre-caching decisions. The controller learns from historical access patterns and feedback about actual data usage to optimize which data to pre-cache, thereby reducing data thrashing while maintaining low latency access.
Solution Approach 2:
The controller performs preliminary actions by pre-fetching and caching data before it is actually requested by accelerator cores. Based on learned access patterns, the controller proactively loads data into the cache memory in advance, so that when the data is needed, it is already available in fast memory, reducing access latency.
2Speed
If the controller pre-caches more data to reduce latency, then data access speed improves, but system resources are wasted due to data thrashing
Solution Approach 1:
The patent applies dynamics by making the pre-caching strategy adaptive and changeable over time. The controller dynamically adjusts which data to pre-cache based on evolving access patterns from multiple accelerator cores. The system transitions from static pre-caching to dynamic, learned-based pre-caching that responds to changing workloads and access behaviors.
Solution Approach 2:
The controller applies local quality by tailoring pre-caching decisions to specific accelerator cores and their individual access patterns. Rather than applying a uniform pre-caching strategy to all data, the controller identifies and pre-caches data that is locally relevant to each accelerator core's workload, improving resource efficiency while maintaining high access speeds.
3Adaptability or versatility
If the storage device manages finite bandwidth for multiple accelerator cores, then resource allocation is controlled, but pre-caching optimization becomes complex
Solution Approach 1:
The patent implements universality by creating a unified pre-caching controller that serves multiple accelerator cores with different workloads and access patterns. The controller learns and adapts to various types of access patterns from different cores, providing a universal solution that handles diverse computational tasks while managing finite bandwidth efficiently across the system.
Data Source
AI summary
A storage device may optimize pre-cache operations based on patterns associated with previously received data requests from accelerator cores in the storage device. The storage device may also optimize pre-cache operations by analyzing the usage of cache data by the accelerator cores. The accelerator cores may perform computational storage functions. The storage device may also include a storage core to receive data requests from an accelerator core to access data stored on a memory device. The storage core may learn a request pattern associated with the data requests and analyze usage of previously cached data by the accelerator core. The storage device may use the request pattern and the usage of previously cached data to optimize its pre-cache operations.