Journal Buffer Rearrangement Across Bank Pools for Faster Reboot
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices face challenges in reducing the open time during reboot due to inefficient management of journal data, which is critical for restoring metadata and user data after unexpected power-offs or crashes.
Innovation Solution
A storage device with a journal buffer and memory controller that temporarily stores journal data in bank pools based on bank and row addresses, rearranging them to optimize access to volatile memory, thereby improving row buffer hit and bank parallelism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If journal data is stored in traditional memory structures, then data storage is simple, but access efficiency and row buffer hit rate are poor
Solution Approach 1:
The memory structure is segmented into multiple bank pools, each containing multiple banks. This segmentation allows parallel access to different journal data entries across banks, improving access speed while maintaining manageable complexity through modular organization.
Solution Approach 2:
The patent introduces a two-dimensional addressing structure with bank addresses and row addresses. This dimensional change enables more efficient data organization and access patterns, allowing the system to achieve better row buffer hit rates by exploiting spatial locality in the expanded address space.
2Loss of time
If journal data is not pre-organized, then storage is faster, but access during reboot requires more time
Solution Approach 1:
Journal data is pre-organized during the write operation by distributing entries across multiple banks and rows based on their addresses. This preliminary organization enables faster access during reboot without requiring complex real-time processing, as the data is already positioned for efficient retrieval.
Solution Approach 2:
The patent changes the organizational parameters of journal data by using dual addressing (bank address and row address) instead of traditional single addressing. This parameter change allows the system to optimize both write and read operations by controlling data placement at multiple hierarchical levels.
3Reliability
If traditional memory access patterns are used, then control is simple, but row buffer hit rate is low
Solution Approach 1:
By segmenting memory into multiple banks within bank pools, the system can service multiple memory requests in parallel. This segmentation increases the probability of row buffer hits by allowing independent access to different banks, effectively reducing control complexity while improving reliability.
Solution Approach 2:
The bank pool structure provides universal access patterns that work for different types of journal data and access scenarios. The same multi-bank architecture handles both sequential and random access patterns efficiently, achieving high row buffer hit rates without requiring specialized control logic for different cases.
Data Source
AI summary
A storage device includes a journal buffer configured to store a plurality of pieces of journal data, the journal buffer including a plurality of bank pools and a journal data region, where each of the plurality of pieces of journal data includes a first data item corresponding to a bank address and a second data item corresponding to a row address and a memory controller configured to control the journal buffer to temporarily store first pieces of journal data from among the plurality of pieces of journal data of which the first data items are equal to each other in at least one of the plurality of bank pools and rearrange the first pieces of journal data based on results of a comparison between the second data items corresponding to the first pieces of journal data.