CXL Memory Pool Allocation With Address Translation for Cost Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory architectures in data centers face inefficiencies due to high memory usage costs when business volumes are low, leading to resource waste and utilization challenges.
Innovation Solution
A resource allocation method and apparatus that dynamically allocates memory resources by selecting a target memory module matching the required capacity, performing address translation, and allocating it to the host, utilizing a memory management unit and allocation unit connected via CXL bus, UART, and high-speed serial bus for real-time monitoring and control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If enough CXL memory modules are connected to meet peak business demands, then memory capacity is sufficient, but memory usage cost increases and utilization rate decreases when business volume is low
Solution Approach 1:
The patent implements dynamic memory allocation where the memory management unit continuously monitors host memory usage and allocates or releases memory modules based on real-time business demands. When business volume is low, excess memory modules are released back to the memory pool, reducing usage cost. When peak demand occurs, memory is dynamically allocated from the pool, ensuring sufficient capacity without permanently deploying all memory modules.
Solution Approach 2:
The patent creates a shared memory pool that serves multiple hosts simultaneously. The memory management unit acts as a universal allocator that can assign memory modules to any host in the pool based on current needs, rather than dedicating specific memory modules to specific hosts. This multi-functional approach allows the same memory resources to serve different hosts at different times, improving overall utilization and reducing total memory requirements.
2Quantity of substance
If enough CXL memory modules are connected to meet peak business demands, then memory capacity is sufficient, but memory resource utilization rate decreases when business volume is low
Solution Approach 1:
The system dynamically adjusts memory allocation based on real-time monitoring of host memory usage. The memory management unit continuously assesses whether hosts need additional memory and whether memory modules should be released back to the pool. This dynamic adjustment ensures that memory resources are actively utilized when needed and properly released when not needed, maintaining high utilization rates across varying business volumes.
Solution Approach 2:
The patent implements a feedback mechanism where the memory management unit monitors host memory usage status and uses this information to make allocation decisions. The system receives feedback about memory consumption patterns and adjusts the memory pool allocation accordingly, releasing memory when hosts have sufficient capacity and allocating when hosts need expansion, thereby optimizing utilization rates.
3Reliability
If memory modules are allocated to hosts, then memory needs are met, but resource allocation efficiency decreases when allocation is not optimized
Solution Approach 1:
The patent pre-establishes a memory pool with multiple memory modules available for allocation. Rather than allocating memory on an ad-hoc basis when needed, the system prepares the memory pool in advance and maintains it ready for rapid allocation. The memory management unit is pre-configured to monitor host needs and can quickly allocate appropriate memory modules from the pre-prepared pool, improving both availability and allocation efficiency.
Solution Approach 2:
The memory management unit serves as an intermediary between hosts and the memory pool, intelligently matching memory allocation requests with suitable available modules. Rather than direct host-to-memory allocation, the management unit mediates the process by monitoring host needs, selecting appropriate memory modules from the pool based on capacity requirements and availability, and managing the allocation process. This intermediary approach optimizes allocation efficiency while ensuring reliable memory availability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the present disclosure provide a resource allocation method and apparatus, an electronic device, and a storage medium, which relate to the field of computer storage technology. In the method, through a memory management unit in a resource allocation system, in response to a memory obtaining request forwarded by a switch in the resource allocation system, a required memory capacity and a memory start address carried by the memory obtaining request are obtained; a target memory module that matches the required memory capacity is selected from at least two memory modules in the resource allocation system based on a memory allocation unit in the resource allocation system; and address translation is performed on the target memory module based on a memory address router in the memory allocation unit and the memory start address, and the target memory module after the address translation is allocated to the target host, ensuring the rapid allocation of free memory modules in the system, improving utilization rate of memory resources and reducing the use cost of the memory resources.