Stand-alone test method, device and system based on NVMe disk array

Abstract

The invention provides a stand-alone test method, device and system based on an NVMe disk array, and the method comprises the following steps: generating test data through software data generation, the test data comprising sending codebook data and receiving codebook data for comparison; using the NVMe disk array to preset and store the sending codebook data, and reading the sending codebook data out from the NVMe disk array during testing and sent to the to-be-tested device; using the NVMe disk array to preset and store codebook data for receiving and comparing, receiving data sent by the to-be-tested device, reading the codebook data for receiving and comparing out from the NVMe disk array and compared with the received data, and verifying the correctness of the received data. According to the invention, the data read-write bandwidth and the data storage capacity are improved, the transmission and comparison of the ultra-large codebook can be realized, the problem of high-bandwidth test of embedded equipment is solved, and the problem of real-time comparison of the complex and irregular large codebook in the test is solved.

Description

technical field [0001] The invention relates to the field of embedded device testing, in particular to a single-machine testing method, device and system based on NVMe disk arrays. Background technique [0002] Before using an embedded device, it is necessary to test and verify the core algorithm processing unit, data transmission unit and management control unit of the device. This type of test usually needs to complete functions such as test data sending, test data receiving analysis and test data storage. [0003] In order to meet the needs of high-speed data processing and large codebook testing, existing testing technologies generally use multi-device parallel processing, which has a low degree of device integration and cannot use super-large real data or codebooks to simulate complex scenarios; in order to meet large-capacity storage Requirements, the existing test technology generally adopts the storage server method, the real-time storage bandwidth is low and an add...

AI Technical Summary

Problems solved by technology

[0003] In order to meet the needs of high-speed data processing and large codebook testing, existing testing technologies generally use multi-device parallel processing, which has a low degree of device integration and cannot use super-large real data or codebooks to simulate complex scenarios; in order to meet large-capacity storage Requirements, the existing test technology generally adopts the storage server method, the real-time storage bandwidth is low and an additional storage server needs to be configured; in order to meet the high-speed data analysis and comparison requirements, the existing test technology generally performs regular codebook sequence analysis and error statistics, Or use the method of storing first and then comparing data for data comparison, which cannot realize real-time comparison processing of real data and large-codebook data

Images