Variable Memory Access Granularity for Fine-Grained Data Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face inefficiencies in data access due to coarse granularity, leading to significant energy consumption when accessing small, dispersed data units, as they often retrieve and overwrite unnecessary data.
Innovation Solution
Implementing a dual-channel memory component with variable access granularity, allowing transition between volume-mode and resolution-mode operations, where resolution-mode reduces data access granularity by a factor of four, utilizing sub-row and sub-column addressing to minimize energy consumption and improve efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If coarse granularity memory access is used, then device complexity is reduced, but energy consumption increases when accessing small data units
Solution Approach 1:
The memory system is divided into multiple banks (e.g., 32 banks) that can be independently accessed. Each bank can operate in different modes (volume-mode or resolution-mode), allowing fine-grained access to specific small data units without activating the entire memory array, thus reducing energy consumption while maintaining manageable complexity through modular organization.
Solution Approach 2:
The memory system dynamically switches between volume-mode and resolution-mode operations based on access requirements. Resolution-mode provides fine-grained access for small data units with reduced energy consumption, while volume-mode handles bulk operations. This dynamic adaptability resolves the contradiction by optimizing the balance between access granularity and energy efficiency for different workloads.
2Use of energy by moving object
If fine-grained data access is implemented, then energy consumption is reduced, but device complexity increases
Solution Approach 1:
The memory is segmented into multiple independently controllable banks, each capable of fine-grained access operations. This segmentation allows the system to achieve fine-grained access capability without proportionally increasing overall system complexity, as each bank operates independently with standardized control logic.
Solution Approach 2:
Each memory bank is designed to be multi-functional, supporting both volume-mode and resolution-mode operations. This universality means that the same hardware structure can adapt to different access granularity requirements, reducing the need for separate complex circuits for different access modes and thereby limiting the increase in device complexity.
3Device complexity
If coarse granularity access is used, then device complexity is lower, but productivity decreases due to unnecessary data retrieval
Solution Approach 1:
The memory system dynamically adapts its access mode based on the size and distribution of data being accessed. For small, dispersed data units, resolution-mode is activated to retrieve only necessary data, improving productivity by eliminating unnecessary data transfer. For bulk operations, volume-mode maintains efficient throughput. This dynamic behavior resolves the contradiction between complexity and productivity by optimizing performance for different access patterns.
Solution Approach 2:
By segmenting the memory into multiple banks with independent control, the system can selectively activate only the necessary banks for each access operation. This segmentation enables efficient handling of small data units by accessing only specific banks rather than retrieving data from the entire memory array, thereby improving productivity without requiring a complete redesign of the memory architecture.
4Use of energy by moving object
If resolution-mode operation is implemented, then energy consumption is reduced, but device complexity increases
Solution Approach 1:
The memory control circuitry is designed with universal control logic that handles both volume-mode and resolution-mode operations through a unified interface. This multi-functionality allows the same control circuits to adapt to different access granularities without requiring separate complex control paths, thereby limiting the increase in device complexity while enabling energy-efficient resolution-mode operation.
Solution Approach 2:
The control circuitry dynamically configures its operation based on the access mode requirement. Control signals are dynamically adjusted to enable either volume-mode or resolution-mode access, allowing the system to achieve fine-grained access capability without permanently increasing hardware complexity. The dynamic reconfiguration uses existing circuit resources efficiently, resolving the contradiction between enhanced functionality and complexity.
Data Source
AI summary
An integrated-circuit memory component receives, as part of respective first and second memory read transactions, a first column access command that identifies a first volume of data and a second column read command that identifies a second volume of data, the second volume of data being constituted by not more than half as many data bits as the first volume of data. In response to receiving the first column access command, the integrated-circuit memory component transmits the first volume of data as N parallel bit-serial data signals over N external signaling links. In response to receiving the second column access command, the integrated-circuit memory component transmits the second volume of data as M parallel bit-serial data signals over M of the N external signaling links, where M is less than N.