How rack-scale flash can further boost high performance health IT

By using the NVMe protocol, rack-scale flash vendors can deliver much lower latency and improve the IOPS of an overall system, making rack-scale flash best suited for high-performance computing needs.

Jeff Rowe | Jan 08, 2018 12:00 am

As with any technology that is at once relatively new and constantly changing, there’s no shortage of terminology and new product that health IT managers need to stay abreast of when it comes to flash data storage.

For example, what exactly is rack-scale flash?  According to recent primer at Tech Target,   “rack-scale flash is flash-only storage that uses a high-speed interface to connect the storage more directly to the CPU than with a traditional storage array. The most common connection is nonvolatile memory express (NVMe) connected via a network fabric.

Another name for rack-scale flash is “shared flash storage,” as it creates a pool of storage that is shared by servers over the high-speed interconnect inside one or more racks.

According to the article, there are two main types of rack-scale flash suppliers: those that use off-the-shelf solid-state drives (SSDs) and those that use custom flash modules. As for the benefits and drawbacks of rack-scale flash, experts say “rack-scale flash boosts the performance of any system, even more than all-flash arrays that don't connect via low-latency NVMe protocols. All-flash array vendors claim IOPS ranging from 200,000 to the low millions.”

On the other hand, perhaps not surprisingly, there is a cost element to rack-scale flash. “Not only is it an all-flash array system, (but) most rack-scale flash vendors use the NVMe-oF protocol, which is still an early stage technology compared to SCSI or TCP/IP. “

Another drawback related to cost is the sheer amount of flash used in rack-scale flash, which has a very high storage density in relation to a standard all-flash array.