Hands-On Testing and Analysis

Notes from the NAND Front


I ran across an interesting article on the site about the state of the NAND flash market. While much of the article covers trends I wrote about in a post at SearchSolidStateStorage like the difficulty vendors have had in ramping up 3D NAND production there were a few tidbits I thought you, my dear readers, would be interested in.

The first is that the flash vendors underestimated demand. According to our friend, and favorite memory analyst, Jim Handy the NAND cabal planned for a 40% rise in demand in 2017 but real demand has been more like 45% higher than last year.

“We are expecting to see shortages until the middle of 2018,” Handy said. “What we are looking at is how long it’s going to take for the last problem to be solved in order to make 3D NAND a cost-effective part to build. Until that happens, we expect the shortages to grow more acute, because demand will continue to grow while production does not.”

3D NAND production across the industry last year was running at a rate of 350,000-400,0000 wafer starts per month. While the folks at Applied Materials that make chip production gear and follow this sort of thing expect that to rise to 500,000-700,000 wafers/mo by the end of 2017. According to Handy, a 2D NAND wafer costs about $1200 to produce while a wafer of 3D chips costs about $2000. Since the 3D NAND wafer holds 17.2TB vs 5.6 TB for the 2D net cost per GB comes to 12¢ for 3D vs 21¢ for planar NAND.

The Chinese are investing billions in a domestic semiconductor industry, including an attempt to break into the NAND market, while the existing players are building out new fabs as fast as their pocketbooks and contractors will let them. Pacific Crest Securities estimates that investment in NAND flash production will rise to almost $24 billion, up 24% from 2016, which is almost a third of the total $76.6 the industry will invest in new plant and equipment for all chips.

The article also discusses the challenges in manufacturing ever denser NAND. While I found much of it interesting most storage folks don’t need to worry about the difficulties of ion drilling holes that are 60 times as deep as they are wide through hundreds of layers of NAND materials.

Note: I wrote this back in June but just noticed (11/11/2017) that it was still sitting as a draft.  Sorry folks.  I’m still hoping the current flash and NAND shortages ease by mid-2018 so I can afford the NVME-oF kit I’ve been drooling over for the lab.