Let’s deconstruct the “flash will die” theory.
I’ve read about a articles recently predicting a sudden and abrupt end to the flash market. The rationale given was that as flash NAND gets cheaper, it also becomes less reliable. Here is due to the decreasing flash lithography and the increasing write density (choice of bits per cell). Each of those lowers the life expectancy of the technology. The concept is that, as this continues to happen, flash turns into unsustainable and the industry will move directly to something else.
While I agree that flash will someday get replaced by another technology, i don’t believe that’s going to happen anytime soon. Except for NVDIMM, most competing memory technology is five to 6 years faraway from being appropriate for the enterprise, and NVDIMM should be used only in small quantities as a result of cost.
There’s various technology that supports the flash NAND market. This includes flash controller technology that controls how and where flash is written. The difficulty with the “flash will die theory” is that it assumes that, because the flash NAND market advances, the encircling market will stand still. On the contrary, we’ve seen at the very least as much innovation in flash controllers as we’ve seen in flash NAND.
[Which flash is ideal for you? See our Quick Guide To Flash Storage Latency Wars.]
Another error on this theory is that it assumes that storage system builders won’t adapt and improve their technology. Again on the contrary, we’ve seen these system designers create technologies that minimize flash writes by leveraging deduplication and compression. We think these vendors to continue improving these technologies and increase their ability to compress and identify redundant data segments.
We also are commencing to see initial implementations of flash tiering, where data is in the beginning written to a small but more reliable single-level cell-based tier, then staged to multi-level cell because it stabilizes and becomes more read-only in nature. Going forward, these vendors could easily create a small SLC front end to behave as a shock absorber to initial data writes, after which leverage increasingly lower price and better density flash NANDs.
The final problem with the “flash will die” theory is that it assumes that flash-based storage systems have far to visit be cost competitive with traditional disk storage systems. It really is simply not the case. Most hybrid and all-flash systems are already claiming price parity with high-performance disk arrays, and in our research at Storage Switzerland, we discover that to be pretty accurate. Given a superb archiving strategy, we’re on the point where using flash for almost all (if not all) of your active data is a reality.
Buying power and influence are rapidly shifting to service providers. Where does that leave enterprise IT? Not on the leading edge, that’s needless to say: Only 19% are increasing both the number and capability of servers, budgets are level or down for 60% and just 12% are using new micro technology. Discover more within the 2014 State Of Server Technology report. (Free registration required.)
More Insights