SSD DRAM Reduction via Referring Table Subgroup Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing storage capacity of SSDs leads to a significant increase in the cost of the required DRAM module, which is typically at least 1/1000 of the total storage capacity, making it an unavoidable expense.
Innovation Solution
A method is introduced that reduces DRAM usage in SSDs by utilizing a referring table in the DRAM module to store physical addresses of subgroups of the mapping table in non-volatile memory units, allowing for efficient access and reducing the need to copy the entire mapping table from flash memory to DRAM, thereby minimizing DRAM size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the storage capacity of SSD is increased, then the storage capacity is improved, but the cost of DRAM module increases significantly
Solution Approach 1:
The mapping table is divided into multiple subgroups, where only the currently needed subgroup is loaded into DRAM while other subgroups remain in flash memory. This segmentation allows the system to handle large storage capacities without proportionally increasing DRAM size, thus reducing cost while maintaining performance.
Solution Approach 2:
Instead of loading the entire mapping table into DRAM, only the necessary portion (current subgroup) is loaded. This partial action approach reduces DRAM requirements significantly while still providing fast access for current operations, addressing the cost issue without sacrificing functionality.
2Speed
If the entire mapping table is copied from flash memory to DRAM, then read/write speed is improved, but DRAM size and cost increase
Solution Approach 1:
The mapping table is segmented into subgroups, and only the active subgroup is copied to DRAM at any given time. This maintains fast read/write speeds for current operations while keeping DRAM size minimal, avoiding the need to allocate large DRAM capacity.
Solution Approach 2:
Only the necessary subgroup of the mapping table is loaded into DRAM rather than the entire table. This partial loading strategy achieves fast access speeds for current operations without requiring excessive DRAM capacity, thus resolving the contradiction between speed and size.
3Reliability
If a small DRAM module is used, then cost is reduced, but access speed to mapping data decreases
Solution Approach 1:
The needed subgroup of the mapping table is pre-loaded into DRAM before it is actually needed for operations. This preliminary action ensures that when read/write operations occur, the data is already in fast DRAM, maintaining high access speeds despite using minimal DRAM capacity.
Solution Approach 2:
By loading only the necessary subgroup into DRAM in advance, the system achieves fast access speeds for current operations without requiring large DRAM capacity. This partial pre-loading strategy balances cost and performance effectively.
4Reliability
If the DRAM module size is reduced, then cost is reduced, but the ability to store complete mapping table is compromised
Solution Approach 1:
The mapping table is divided into subgroups stored in flash memory, with only the active subgroup in DRAM. This segmentation allows the complete mapping table to be maintained in flash (inexpensively) while keeping DRAM minimal, resolving the contradiction between storage capacity and cost.
Solution Approach 2:
Flash memory acts as an intermediary storage layer that holds the complete mapping table when not in use. This allows DRAM to be kept small while still providing access to the full mapping table through the flash memory intermediary, balancing cost and capacity requirements.
Data Source
AI summary
A SSD and a method for reducing use of DRAM in the SSD are disclosed. The method includes the steps of: A. providing a referring table in a DRAM module of a SSD; B. providing a logical-to-physical address table in the DRAM module; C. receiving a command for accessing a target data in a target logical address of the SSD; D. checking if one physical address is stored in the logical-to-physical address table; E. executing the command by using the mapping data in the subgroup or copying a corresponding subgroup including one mapping data for the target logical address from the mapping table to the DRAM module via the referring table; and; and F. adding a target physical address of the DRAM module where the mapping data for the target logical address is stored to the logical-to-physical address table so that the target logical address is able to correspond thereto.


