Variable Burst Length Page Size for MRAM Energy Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic random access memories (MRAMs) face higher energy consumption rates compared to non-magnetic memory devices like DRAMs, especially when accessing large pages, which increases power usage and inefficiency.
Innovation Solution
Configuring memory devices with smaller page sizes, such as 8 bytes per page, and allowing burst lengths to be defined based on the number of bytes in a read or write command, enabling flexible activation of memory arrays and auto-precharge commands to minimize energy consumption and improve reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If large page sizes are used to minimize page refreshing, then memory access efficiency is improved, but energy consumption increases
Solution Approach 1:
The patent implements dynamic page size configuration where the memory device can be programmed to operate with different page sizes (e.g., 8 bytes, 16 bytes, 32 bytes, or 64 bytes) based on the specific application requirements. This allows the system to adaptively select optimal page sizes that balance between access efficiency and energy consumption, rather than being fixed to large page sizes.
Solution Approach 2:
The invention changes the parameter of page size from a fixed large value to a configurable variable. By modifying the page size parameter according to actual access patterns and performance requirements, the system achieves better energy efficiency while maintaining necessary access efficiency, directly resolving the contradiction between these two parameters.
2Ease of operation
If self referencing reads are performed to enable data reading, then read capability is achieved, but energy consumption increases due to bit switching
Solution Approach 1:
The patent extracts and eliminates unnecessary bit switching operations from the read process. By using conventional read operations that directly sense memory cell states without requiring full bit switching, the invention removes the harmful energy-consuming aspect while preserving the essential read capability.
Solution Approach 2:
The invention employs simpler, lower-energy read mechanisms that consume less energy per operation. Instead of expensive self-referencing reads that switch all bits, the patent uses more efficient sensing methods that achieve the same read capability with significantly reduced energy expenditure.
3Productivity
If page size is increased to reduce refreshing operations, then access frequency is reduced, but energy consumption rate increases
Solution Approach 1:
The patent implements dynamic adjustment of page size based on workload characteristics and access patterns. When access frequency requirements are high, smaller page sizes are used to reduce energy consumption. When batch operations are performed, larger page sizes can be utilized. This dynamic adaptation resolves the contradiction by allowing the system to optimize for energy efficiency during frequent access scenarios.
Solution Approach 2:
The invention changes the page size parameter from fixed to variable, enabling the system to select appropriate page sizes that minimize energy consumption rates while maintaining necessary access frequencies. This parameter flexibility directly addresses the trade-off between productivity and energy loss.
Data Source
AI summary
In some examples, a memory device is configured with a reduced command set and a variable burst length. In some instances, the variable burst length defines a page size associated with data to be loaded into a cache. In other instances, the variable burst length may be set on the fly per read/write command and, in some cases, the burst length may be utilized to define the page size associated with the read/write command.