Interval Tree Data Structure for Near-Storage Memory Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of modern memory systems, with multiple layers of storage, leads to delays in data access as conventional methods sequentially search processor caches, main memory, and persistent storage, resulting in cumulative time delays.
Innovation Solution
A scalable interval tree data structure is used to track the location of data elements across various memory layers, allowing direct access to the data's current storage location, reducing the need for sequential searches and optimizing data retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If sequential search methods are used to access data across multiple memory layers, then data can be retrieved from any storage location, but data access time increases due to cumulative delays
Solution Approach 1:
The system performs preliminary actions by pre-tracking and pre-locating data elements in a data structure before access is needed. This allows the system to know in advance where data is stored across memory layers, enabling direct access without sequential searching and thus reducing access time despite the complex multi-layer memory architecture.
Solution Approach 2:
The patent introduces an intermediary data structure that mediates between the data access request and the complex multi-layer memory system. This intermediary layer tracks data locations and enables direct routing to the correct storage location, eliminating the need to sequentially search through processor caches, main memory, and persistent storage, thereby reducing access time.
2Quantity of substance
If multiple memory layers are implemented to increase storage capacity, then more data can be stored, but data access becomes slower due to sequential search requirements
Solution Approach 1:
The system performs preliminary tracking of data locations across all memory layers and stores this information in a data structure. When data access is needed, the system uses this pre-acquired location information to directly access the data without sequential searching, thus maintaining fast access times even with increased storage capacity from multiple memory layers.
Solution Approach 2:
The patent introduces an intermediary data structure that mediates between data access requests and the multi-layer memory system. This intermediary tracks where data is stored across processor caches, main memory, and persistent storage, enabling direct routing to the correct location and eliminating sequential search delays, thus preserving fast access times while providing expanded storage capacity.
3Productivity
If conventional sequential search methods are used, then data access is simple to implement, but system performance decreases due to cumulative time delays
Solution Approach 1:
The system performs preliminary tracking of data locations and stores this information in a data structure before access is needed. This preliminary action enables fast direct access to data without sequential searching, significantly improving system performance. The complexity of maintaining this tracking structure is offset by the substantial performance gains from eliminating sequential search delays.
Solution Approach 2:
The patent introduces an intermediary data structure that mediates between access requests and the memory system. This intermediary maintains tracking information about data locations across multiple memory layers, enabling direct routing to the correct location. While this adds some structural complexity, it eliminates sequential search delays and dramatically improves system performance and productivity.
Data Source
AI summary
A system is disclosed. The system may include a processor, a first memory connected to the processor, and a second memory connected to the processor. A data structure may include an entry, which may identify that a data is stored in a location. The location may include one of the first memory or the second memory.