Author: Edward Sharp

“May you live in interesting times.” If you haven’t noticed, IT is already there, with cloud, mobile, big data, in-memory, NoSQL, Remote Direct Memory Access (RDMA), Shingled Magnetic Recording (SMR) drives, Non-Volatile Memory (NVM), and the list goes on. Our landscape is undergoing massive disruption as new technologies and techniques enable customers to do more with less. This disruption is a threat to existing businesses, and an opportunity for the fast, the focused, and the bold.

For those of us in the storage business, there is no bigger disruption than NVM. After decades of hard disk drives (HDDs) owning the storage media layer, getting slower (particularly on a per-GB-basis), fatter and cheaper each year, along came flash solid state drives (SSDs) to change everything:  milliseconds of latency improved by orders of magnitude; racks of spindles for performance replaced with a shelf of SSDs; power, heat and noise reduced significantly; and incumbents disrupted by start-ups.

But NVM is memory, and we’ve just been talking about NVM as storage! As my colleague Stephen Bates has noted before, “Let’s stop thinking about NVM as fast storage and start thinking about it as (slow) memory.”  In this new world, computer architectures can be re-imagined, with data available in persistent memory right off the CPU. This isn’t science fiction. The bright folks at MIT have recently demonstrated 20 servers with flash memory performing as well as 40 servers with DRAM. This is a harbinger of things to come.

This will have profound implications for what customers get for the money they put into their infrastructure. However, it requires changes throughout the stack, from applications to devices. Let’s get to work!