Memory System Dynamic Interruption Threshold for Transfer Type
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In memory systems using SSDs, random data transfers lead to increased latency and overhead, necessitating a reduction in the number of commands or the time-out interval to improve performance.
Innovation Solution
A memory system with a command receiving unit, a command queue, a transfer executing unit, an end number counter, an interruption generating unit, and a transfer type conjecturing unit that dynamically adjusts the end number threshold and time-out interval based on whether the transfer type is sequential or random, optimizing interruption timing to reduce host processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If the number of commands to be dealt with together is increased to reduce interruption overhead, then host processing overhead is reduced, but latency increases for random transfer operations
Solution Approach 1:
The patent dynamically adjusts the end number threshold based on transfer type (sequential vs. random). For sequential transfers, a higher threshold is used to batch more commands and reduce overhead. For random transfers, a lower threshold is used to minimize latency by processing commands more quickly. This dynamic adaptation resolves the contradiction by optimizing the threshold for each transfer type's specific requirements.
Solution Approach 2:
The patent changes the parameter (end number threshold) based on the transfer type. By identifying whether transfers are sequential or random, the system adjusts the threshold parameter accordingly - increasing it for sequential transfers to reduce overhead and decreasing it for random transfers to reduce latency. This parameter change strategy directly addresses the technical contradiction.
2Loss of time
If the time-out interval from command receipt to interruption generation is reduced to improve random transfer latency, then responsiveness improves, but the number of interruptions increases reducing throughput
Solution Approach 1:
The patent dynamically adjusts the time-out interval based on transfer type. For random transfers, a shorter time-out interval is set to improve responsiveness and reduce latency. For sequential transfers, a longer time-out interval is acceptable since batching is more beneficial. This dynamic adjustment resolves the contradiction between responsiveness and throughput.
Solution Approach 2:
The patent changes the time-out interval parameter according to the transfer type detected. By setting different time-out values for sequential versus random transfers, the system optimizes both latency and throughput - using shorter intervals for random transfers to improve responsiveness while maintaining longer intervals for sequential transfers to maximize throughput.
3Device complexity
If a fixed end number threshold is used for all transfer types, then system complexity is reduced, but performance is suboptimal for both sequential and random transfers
Solution Approach 1:
The patent applies different end number thresholds for different transfer types (sequential vs. random). Instead of using a single uniform threshold, the system tailors the threshold to the specific characteristics of each transfer type, optimizing performance for both cases. This local quality approach resolves the contradiction by allowing specialized optimization without excessive complexity.
Solution Approach 2:
The patent changes the end number threshold parameter based on the transfer type. By detecting whether transfers are sequential or random and adjusting the threshold accordingly, the system achieves optimal performance for both transfer types. The complexity introduced is minimal - primarily involving transfer type detection and conditional parameter adjustment.
Data Source
AI summary
According to one embodiment, a memory system is provided wherein an interruption generating unit generates an interruption signal for one or more commands executed by a transfer executing unit when an end number counter is greater than or equal to a first threshold. A transfer type conjecturing unit determines whether the transfer type of a first command to be executed after transmitting the interruption signal is sequential transfer or random transfer and sets the first threshold at a value different between when determining being the sequential transfer and when determining being the random transfer.