Persistent Region Summaries in NVRAM Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current database systems waste resources by processing and filtering out rows that do not contribute to query results, as region summaries are volatile and lost when storage devices power off, requiring regeneration and inefficiently updated in block mode storage devices.
Innovation Solution
Storing access-derived region summaries in non-volatile random access memory (NVRAM) to persistently maintain summary data, allowing efficient generation and updating without the high latency of block mode storage, and ensuring coherency during data block writes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If region summaries are stored in volatile memory, then access speed is fast, but data is lost when storage device powers off requiring regeneration
Solution Approach 1:
The patent segments the memory storage system into two parts: volatile memory for frequently accessed region summaries and non-volatile memory for persistent storage. This segmentation allows the system to maintain fast access speeds for active summaries while ensuring data persistence through non-volatile storage, resolving the contradiction between speed and reliability.
Solution Approach 2:
The patent introduces non-volatile memory as an intermediary between volatile memory and permanent storage. This intermediary layer maintains region summaries persistently, allowing the system to recover summaries after power loss without regeneration, thus solving the data persistence problem while maintaining fast access characteristics.
2Reliability
If region summaries are updated in block mode storage devices, then data is persisted, but high latency overhead occurs
Solution Approach 1:
The patent segments the storage update path into two channels: non-volatile memory for summary updates and block mode storage for actual data blocks. This segmentation allows summaries to be updated with low latency in non-volatile memory while data persistence is achieved through the block mode storage, eliminating the high latency overhead.
Solution Approach 2:
The patent extracts the region summary update operation from the block mode storage path and places it in non-volatile memory. This extraction removes the high latency bottleneck of block mode storage from the summary update path, allowing fast updates while maintaining data persistence through the non-volatile memory's inherent persistence capabilities.
3Manufacturing precision
If filtered data block scans are performed on all regions, then complete filtering is achieved, but network and memory resources are wasted on rows that will be filtered out
Solution Approach 1:
The patent performs preliminary filtering using region summaries stored in non-volatile memory before executing the actual filtered data block scan. By checking the summaries first, the system can identify and skip regions that will definitely be filtered out, thus achieving complete filtering while avoiding the waste of network and memory resources on rows that will not contribute to the result set.
Data Source
AI summary
Region summaries of database data are stored in persistent memory of a storage cell. Because the region summaries are stored in persistent memory, when a storage cell is powered off and data in volatile memory is not retained, region summaries are nevertheless preserved in persistent memory. When the storage cell comes online, the region summaries already exist and may be used without the delay attendant to regenerating the region summaries stored in volatile memory.