Client-Directed Storage Addressing for Lower-Latency Data Writes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing data writing process in storage systems is complex, leading to increased delays and reduced efficiency due to the need for CPUs in storage nodes to determine hard disk logical block addresses and physical addresses, which complicates the data writing process.
Innovation Solution
The client determines the storage address for data without using logical block addresses, directly sending the data to storage devices, thereby reducing the need for CPU involvement and simplifying the writing process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the CPU determines hard disk LBA and the hard disk determines physical address based on mapping relationships, then data can be written to the correct location, but the data writing process becomes complex and writing efficiency is reduced
Solution Approach 1:
The patent extracts the address mapping determination function from the storage node's CPU and hard disk controller, relocating it to the client. The client directly determines the storage address based on the data to be written and sends both the data and address to the storage device, eliminating the need for the storage node's CPU to perform LBA determination and physical address mapping.
Solution Approach 2:
The client performs self-service by autonomously determining the storage address for outgoing data without requiring the storage node's CPU to intervene. The client independently calculates the target address and directly transmits data to the correct storage location, making the storage node's CPU idle during the writing process.
2Measurement precision
If the CPU is involved in determining hard disk LBA and physical addresses, then accurate data placement is achieved, but data writing delay is increased
Solution Approach 1:
The client performs preliminary action by determining the storage address before data transmission. The address calculation is completed in advance at the client side, allowing the storage device to directly write data to the predetermined location without waiting for the storage node's CPU to perform address determination, thereby reducing writing delay.
3Ease of operation
If the storage node's CPU processes address mapping, then data can be routed correctly, but the writing process complexity increases
Solution Approach 1:
The patent extracts the address mapping logic from the storage node's CPU and relocates it to the client. This eliminates the need for complex address translation processes at the storage node, simplifying the writing process while maintaining correct data routing through client-side address determination.
Data Source
AI summary
A storage system, a data processing method, an apparatus, a node, and a storage medium, where a client determines an address that is in a storage unit and that is used to store to-be-written data, and sends the to-be-written data to a first storage device that is in a storage node and that is corresponding to the storage unit. The first storage device stores the to-be-written data without the storage node determining a hard disk LBA corresponding to virtual address space, and without a hard disk determining a corresponding physical address based on the hard disk LBA.