Intel and Micron introduced a revolutionary storage technology called 3D XPoint memory, 1000 times faster than NAND SSD, 1000 times better endurance, Terabyte level density, and with byte level access, this is disruptive and changes the fundamental computer architecture we got used to, now applications and storage stacks need to figure out how to maximize its value.
Having a sub-microsecond non-volatile storage media, WoW !, this is couple of magnitudes faster than common network stack latency, and a magnitude faster than the time current OS storage stacks take to process an IO operation.
Still, not quite DRAM speeds, interfaces, and access model means its not a plug-in replacement for memory.
I assume some will use it as an internal layer in their 500 microsecond FC attached AFA (All Flash Array) hoping for some minor improvement, pretty much the same way legacy storage systems started using SSDs as cache until those brand new AFAs arrived. But to leverage the full potential we need to re-think application stacks and their relations to persistent storage.
To take advantage of sub-microsecond storage we may need to bind applications to low-latency data access libraries which can abstract the new access models along with the traditional disks/file/memory interfaces, and we would need to rethink how we use CPU threads and avoid code blocking and locking. Well with such devices we may need to bypass OS kernel overhead altogether, like the high-performance networking and HPC guys, now slow disks are no longer an excuse.