Intel's 3D NAND has 32 layers and 256Gb per die


— 7:28 PM on November 20, 2014

During an investor webcast this afternoon, Intel revealed that it will offer SSDs based on 3D NAND in the second half of next year. The three-dimensional tech is the product of the firm's joint flash venture with Micron. It stacks 32 planar layers to deliver 256Gb (32GB) of storage in a single MLC die. The 3D NAND can also be packed with three bits per cell to hit 384Gb (48GB) per die.

According to Rob Crooke, senior VP and general manager of Intel's non-volatile memory group, the 3D NAND will enable 10TB SSDs "within the next couple of years." The technology can also squeeze 1TB of storage into a mobile-friendly form factor just two millimeters thick.

Crooke characterized the 3D tech as having a "breakthrough cost" but didn't provide more specifics. He did, however, suggest that Intel may not manufacture the 3D NAND itself. "We have the ability," Crooke said, but Intel will only bring production in-house "if it makes sense." Benefiting from the cost breakthrough presumably requires a substantial capital investment up front.

The 3D NAND's planar layers are built with a coarser fabrication process than the latest and greatest 2D flash. Those layers are pierced by four billion "pillars" that run vertically through the die, but Intel isn't ready to disclose specifics about the underlying process geometry.

Crooke effectively demoed working silicon by running his presentation off a prototype drive. Intel hasn't decided which market segment will get the first taste, though. Datacenters, corporate clients, and PC enthusiasts top the list.

Samsung's 32-layer V-NAND has 86Gb per die in MLC mode and 128Gb in a TLC config, giving Intel and Micron a substantial density advantage—at least per die. That said, a new generation of V-NAND should be ready by the second half of 2015. It will be interesting to see how those chips stack up against the 256Gb monsters Intel has cooked up with Micron.

   
Register
Tip: You can use the A/Z keys to walk threads.
View options

This discussion is now closed.