Weighted Memory-Space Access for Fair Bandwidth Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face issues where certain host systems or applications monopolize bandwidth, leading to reduced performance or complete inability to access data, necessitating mechanisms to ensure fair and minimum throughput access across multiple memory spaces.
Innovation Solution
Implementing a system where a set of relative priorities are assigned to the memory spaces in a storage device, where the assigned priorities of the memory spaces are used to implement a pattern for accessing the memory spaces that is weighted in accordance with the priorities while ensuring each of the memory spaces can be accessed in accordance with a minimum throughput threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If bandwidth is allocated to high-priority memory spaces, then access speed for critical applications is improved, but lower-priority applications may experience reduced access or complete denial of service
Solution Approach 1:
The system dynamically adjusts bandwidth allocation between memory spaces based on priority levels and current access patterns. The controller modifies allocation ratios in real-time to balance between serving high-priority applications quickly and ensuring minimum access for lower-priority applications, resolving the contradiction between speed and reliability
Solution Approach 2:
The system changes the parameter of bandwidth allocation ratios dynamically. By adjusting the proportion of bandwidth assigned to different memory spaces based on priority weights, the system optimizes access speed for critical applications while maintaining minimum service levels for others, thus resolving the speed-reliability tradeoff
2Productivity
If a single application monopolizes bandwidth, then its performance is maximized, but other applications cannot access data
Solution Approach 1:
The system segments the total bandwidth into multiple allocated portions based on priority weights assigned to different memory spaces. This segmentation prevents any single application from monopolizing the entire bandwidth while ensuring each application receives adequate resources for its operational needs, thus maintaining both high performance and multi-application adaptability
Solution Approach 2:
The bandwidth allocation is dynamically adjusted based on the number and priority of active applications. When multiple applications are accessing data, the controller distributes bandwidth proportionally according to assigned weights, preventing monopolization while maintaining optimal performance for each application
3Ease of operation
If weighted interleaved round robin algorithm is implemented, then fair access across memory spaces is achieved, but system complexity increases
Solution Approach 1:
The system implements a periodic weighted interleaved round robin scheduling algorithm that cycles through memory spaces in a predetermined pattern based on priority weights. This periodic approach ensures fair and predictable access distribution across all memory spaces while using a deterministic algorithm that reduces control complexity compared to fully dynamic scheduling mechanisms
Data Source
AI summary
Methods, systems, and devices for weighted distributed-access across memory spaces are described. Multiple commands may be received. The multiple commands may include first commands associated with a first memory space of a memory system that is assigned a first priority of multiple priorities, second commands associated with a second memory space of the memory system that is assigned a second priority of the multiple priorities, and third commands associated with a third memory space of the memory system that is assigned a third priority of the multiple priorities. The multiple commands may be executed using an interleaving pattern that is based on the priorities of the memory spaces.