Supercharge Your Innovation With Domain-Expert AI Agents!

Multi-core CPU (Central Processing Unit) load balancing method, multi-core CPU and solid state disk

A solid-state hard disk, load balancing technology, applied in multi-programming devices, input/output to record carrier, resource allocation, etc., can solve complex algorithms, reduce development efficiency, increase load balancing code development cycle and other problems

Active Publication Date: 2014-10-15
GIGADEVICE SEMICON (BEIJING) INC
View PDF4 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the process of processing read and write requests according to the above scheme, data sharing is likely to be performed between CPUs, that is, in many cases, communication needs to be carried out between two or more CPUs, and between CPUs Communication often requires complex algorithms, which undoubtedly increases the development cycle of load balancing code and reduces development efficiency

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-core CPU (Central Processing Unit) load balancing method, multi-core CPU and solid state disk
  • Multi-core CPU (Central Processing Unit) load balancing method, multi-core CPU and solid state disk
  • Multi-core CPU (Central Processing Unit) load balancing method, multi-core CPU and solid state disk

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] refer to figure 1 , shows a flow chart of Embodiment 1 of a multi-core CPU load balancing method of the present invention, the multi-core CPU may specifically include a front-end CPU and at least one back-end CPU, each of which is used to manage solid-state The flash memory of the hard disk, and the flash memory managed by each back-end CPU is different, and the method may specifically include:

[0037] Step 101, the front-end CPU splits the read and write request from the host into sub-requests with the logical block address LBA as the unit;

[0038] The embodiment of the present invention can be applied to a solid-state hard disk device that uses a multi-core CPU to implement FTL, and is used to improve the development efficiency of the multi-core CPU load balancing code on the device.

[0039] Assuming that the size of each LBA is 4096bytes, and the read and write request involves reading 24 LBAs starting from 0, then the front-end CPU can split it into 24 sub-reque...

Embodiment 2

[0051] refer to figure 2 , shows a flow chart of Embodiment 2 of a multi-core CPU load balancing method of the present invention, the multi-core CPU may specifically include a front-end CPU and at least one back-end CPU, each of which is used to manage solid-state The flash memory of the hard disk, and the flash memory managed by each back-end CPU is different, and the method may specifically include:

[0052] Step 201, the front-end CPU splits the read and write request from the host into sub-requests with the logical block address LBA as the unit;

[0053] Step 202, the front-end CPU obtains the mapping relationship between the sub-request and the corresponding back-end CPU according to the number of back-end CPUs and the number of LBAs contained in each flash page; in the mapping relationship, each sub-request corresponds to a unique back-end CPU;

[0054] Step 203, the front-end CPU sends the sub-request to the corresponding back-end CPU according to the mapping relation...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-core CPU (Central Processing Unit) load balancing method, a multi-core CPU and a solid state disk. The multi-core CPU comprises a front end CPU and at least one back end CPU, wherein each back end CPU is used for managing a flash memory of the solid state disk, and flash memories which are managed by different back end CPUs are different. The method comprises the steps of decomposing a read and write request from a host into sub-requests which take logical block addresses (LBA) as units by the front end CPU; obtaining the mapping relation between each sub-request and the corresponding back end CPU by the front end CPU according to the number of the back end CPUs and the number of LBAs which are contained by each flash memory page, wherein in the mapping relation, each sub-request corresponds to the unique back end CPU; sending the sub-requests to the corresponding back end CPUs by the front end CPU according to the mapping relation, and carrying out flash translation layer (FTL) processing on the sub-requests in the managed flash memories by the back end CPUs. According to the multi-core CPU load balancing method, the multi-core CPU and the solid state disk, the software design is simplified, and the development efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of data storage, in particular to a multi-core CPU load balancing method, a multi-core CPU and a solid-state hard disk. Background technique [0002] SSD (Solid State Disk) is a type of hard disk that uses flash memory as a storage medium. Compared with traditional hard disks, it has the advantages of higher speed, lower power consumption, and lower noise. [0003] Since the flash memory used in solid-state drives has a limit on the number of erasing and writing, once a certain flash memory block is erased / written for a certain number of times, data may not be written or data may be damaged. In order to avoid this situation, the solid-state hard drive introduces FTL (Flash translation layer), FTL is mapped to the PBA (Physical Block Address, Physics Block Address) of the flash memory through the LBA (Logical Block Address, Logic Block Address) of the host , to manage each physical block of the flash memory,...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F3/06G06F9/50
Inventor 曹堪宇朱荣臻高会娟
Owner GIGADEVICE SEMICON (BEIJING) INC
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More