How to solve the 0e reliability problem of domestic SSD?

Recently, a fan sent a private message in the background asking about the 0e issue of process NAND. I feel it is necessary to share with you some of the editor’s opinions and precautions related to SSD purchase.

To be honest, when the editor saw the word 0e, I was a little confused for a while. What is 0e? I also searched it and found that there are quite a few. Mainly entries from Yangtze Memory and Samsung.

When I clicked on it, I discovered that 0e is Media Error. Forgive me for being ignorant.

In the nvme protocol specification, this project defines: The increase in Media Error count is mainly caused by E2E, NAND media UECC, CRC, etc. Our common ones are basically NAND media UECC.

Speaking of Media Error, we must mention the definition of UBER in the JEDEC specification.

SSD UBER (Uncorrectable Bit Error Rate) refers to the ratio of the number of data errors per bit read to the total number of reads that is still generated by the solid state drive (SSD) after applying any specific error correction mechanism.

SSD UBER is an important parameter describing SSD hard disk and is used to measure the stability and reliability of SSD. It reflects the probability that the SSD will have irreparable data errors during the reading process, that is, the read error rate of the SSD. The UBER of enterprise-grade SSD is an order of magnitude higher than that of consumer-grade UEBR. At present, manufacturers can basically achieve 10^-17 or higher, and some SSD products can reach 10^-19.

Guess you like

Origin blog.csdn.net/zhuzongpeng/article/details/132769009