Dual-Node Flash Array Data Access Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The management of accessing Flash memory in all-flash array (AFA) servers is complicated due to differences between SSDs and HDDs, leading to degraded performance, as existing control methods may become improper or invalid.
Innovation Solution
A dual-node architecture where one node acts as a master and the other as a slave, with data being stored in DRAM and then notified to be written to SSD only after successful storage in both nodes, ensuring proper data management and synchronization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing control methods are used for SSD management in AFA servers, then data storage functionality is provided, but system performance is degraded and access control becomes complicated
Solution Approach 1:
The system is divided into master node and slave node segments, each with dedicated DRAM caches. The master node handles data reception and initiation, while the slave node provides verification and notification, segmenting the complex SSD access control into manageable functional units that operate semi-independently
Solution Approach 2:
DRAM caches are introduced as intermediary buffers between the network interface and SSDs. Data flows through DRAM before being written to SSD, and the dual-node DRAM architecture acts as an intermediary verification layer that simplifies access control by decoupling the network interface from direct SSD management
2Speed
If data is written directly to SSD upon receipt, then storage speed is fast, but data integrity and protection are compromised
Solution Approach 1:
The system performs preliminary actions by storing data in DRAM caches of both master and slave nodes before committing to SSD writing. The slave node must successfully store data in its DRAM and send a notification back to the master node before the master proceeds with SSD writing, ensuring preliminary verification of data integrity
Solution Approach 2:
Dual DRAM caches in master and slave nodes serve as cushioning buffers that protect against data loss. If the master node fails before SSD writing, the slave node's DRAM contains a verified copy. This beforehand cushioning mechanism ensures data protection without significantly impacting writing speed
Data Source
AI summary
The present invention provides a server including a SSD, a first node and a second node, wherein the first node comprises a first processor and a first memory, and the second node comprises a second processor and a second memory. When the first processor receives data from another device via network, the first processor stores the data in the first memory, and the first processor further sends the data to the second node; when the second processor receives the data from the first node, the second processor stores the data in the second memory, and the second processor further sends a notification to the first node to inform that the data is successfully stored in the second memory; and after and only after the first processor receives the notification from the second node, the first processor starts to write the data into the SSD.