[ad_1]
Nearly a year after it was first announced, Solidigm’s 192-layer, 61.44TB PCIe 4.0 NVMe SSD is about to get some serious competition from Samsung.
The South Korean tech giant has quietly unveiled the BM1743, a 176-layer, 61.44TB SSD, created using its 7th-generation QLC V-NAND technology. This nearly doubles the layer count from previous iterations and enhances bit density while lowering costs.
Although the BM1743 will initially be made available in a U.2 form factor (2.5-inch), Samsung says there will be also be an E3.S variant to follow that will support PCIe Gen5 for an additional performance boost.
Improved reliability
The BM1743 promises sequential read speeds of up to 7200 MB/s and write speeds of up to 2000 MB/s. Random read and write speeds have quadrupled compared to Samsung’s previous BM1733 SSD, released in 2020. Delivering a giant leap from its predecessor’s 15.36TB, the BM1743’s capacity has soared to 61.44TB, with Samsung’s roadmap suggesting an even larger 122.88TB model is on the cards.
Samsung’s BM1743 not only increases storage capacity and data transfer rates but also shows reduced power consumption and improved reliability. The drive writes per day (DWPD) metric has increased from 0.18 to 0.26, and the power-off data retention period has extended from one month to three months.
Naturally, the new drive isn’t aimed at consumer PCs but is intended for use in data centers, reducing costs and boosting efficiency in data-heavy operations, benefiting cloud services and AI workloads.
Samsung and Solidigm aren’t the only manufacturers promising large-capacity SSDs. Western Digital also recently announced a 64TB SSD, which aims to address data storage for AI applications. Part of WD’s expanded Ultrastar DC SN655 enterprise-class range, this was announced at Computex 2024 and is expected to arrive later this year. It will likely only be a matter of time before the likes of SK hynix, Micron and Kioxia follow suit and debut their own similarly sized drives.
More from TechRadar Pro
[ad_2]
Source link
