Intel plans to ship 3D NAND flash chips next year that will allow it to cram more bits into solid-state storage.
Its 3D NAND will have twice the density of competing products on the market now, Intel claims. Samsung, a key rival, is already on its second generation of SSDs built with 3D technology.
3D NAND has multiple layers of transistors stacked on top of each other in a cube. Intel’s chips will have 32 layers. Samsung is shipping SSDs made with 32-layer flash, but Intel says its products will hold twice as many bits: 256 billion bits on a single die using MLC (multilevel cell), the most common form of flash.
Flash storage is much faster and more power-efficient than spinning disks but remains more expensive to manufacture. By doubling the capacity of a single die, Intel thinks it can achieve a breakthrough in the cost of flash so it can go into a wider range of systems. Products that already use flash will be able to have more storage for a given price.
For the high end of the market, Intel will be able to fit 1TB of data on a NAND chip just two millimeters thick, said Rob Crooke, vice president and general manager of Intel’s Non-Volatile Memory Solutions Group. He announced the plans at Intel’s investor meeting on Thursday.
“We’re no longer having to make a compromise on how much storage we can fit in there,” Crooke said. Within the next two years, Intel plans to build storage cards for enterprise servers that have more than 10TB of capacity.
The technology also increases what Intel can do for less expensive systems.
“You can [have] good-enough storage, so something less than a terabyte, at a much, much lower cost,” he said. For thin and light devices such as two-in-one tablet-PCs, “you can get as much storage as you want,” he said.
Flash still makes up only about 20 percent of storage sold today, averaged across all types of systems, Crooke said. Current projections see flash reaching 50 percent of the notebook market and 35 percent of the server market by 2018, he said.