Does an all-flash data centre make business sense?
Share on Twitter.
Get the most reliable SMTP service for your business. You wished you got it sooner!
September 8, 2014
Here's a good and intriguing question some in the IT industry are currently asking themselves: does an
all-flash data centre make economic sense?
But if you look at this in a blunt manner, the hardware and software components may be there,
but what about the business case, including the overall price/performance and total cost of ownership?
Does it make economic sense?
Of course, such a data centre would be free of power-gobbling and rack space-consuming
spinning disk enclosures (SANs), bringing economic savings in cooling abd power, but storing
data bits in NAND costs more than writing them onto disk-platter recording media.
Unless the price of NAND technology comes down a lot, the jury is still out on that question.
But let's look at where flash can realistically substitute for traditional hard drives in data centres
and where it cannot, moving from high-access rate and latency-sensitive applications to low-access
With data that is of high value and needed very fast, flash is replacing performance disk – 15,000 rpm and
increasingly 10,000 rpm drives at the moment. It is starting to appear as DIMM-connected flash, using
the CPU memory bus and providing the lowest latency access outside of DRAM, meaning roughly 5 to 10 µs
To be sure, NAND is also appearing on PCIe-connected flash cards, with Fusion-io as the perceived
Access latency is in the 75 µs area, say seven times slower than flash DIMMs. A server using
conventional hard drives instead of DIMMs or PCIe flash would have data-access latency of 5 to 10 milliseconds,
which is about 1,000 times longer than DIMM flash and 100 times longer than PCIe flash.
Disk-data access latency is the time needed for a disk's read/write head to move across the platter's
surface to the right track and then for the right section of the track to move under the head.
Unsurprisingly, this waiting for mechanical events takes almost an eternity when compared with solid-state data access.
There are a few software products that turn server-attached flash into storage memory and so help
avoid data passing through the host operating system's disk-based IO subsystem, speeding up data access.
Where some very critical applications merit, such as stock or futures trading looking at
arbitrage opportunities, require extremely fast compute times then avoiding latency delays is
crucial and the cost of the flash is easily justifiable.
How about where a group of servers need a shared storage resource? Can that be done like a virtual SAN
from the individual servers' flash storage? That's another interesting question that needs to be
However, none of the hyper-converged server/storage/system appliance vendors do this currently. You may
be able to put together various HP Proliant server configurations with the P-4000 StoreVirtual VSA.
And in the same manner, you could possibly use Atlantis USXs with an all-flash server config and
build a cross-server flash storage pool.
However, these are not typical systems per se. Generally speaking, no all-flash clustered server
storage systems are available. But you can build them, or have them built, possibly by an HP or Atlantis
Life becomes easier if you need a networked all-flash array. There are at least fourteen suppliers
in the U.S. with products ranging from newly designed arrays to older arrays in an all-flash configuration.
The choice is yours.
Startups such as Pure Storage and SolidFire have arrays designed from the ground up which are
relatively light on data management services.
Mainstream suppliers such as Dell, HDS, HP and NetApp have developed all-flash versions of their
existing arrays which inherit the arrays’ data management services.
For example, NetApp's EF-Series is an all-flash array based on the E-Series and you can also select
all-flash versions of the Data ONTAP FAS arrays which provide all of ONTAP's data management services.
Although all-flash mainstream arrays are possible, most deployments of the arrays are hybrid, using both
flash and disk, because disk access is slower, but naturally, flash access costs more.
Source: ITN 2020.
Get the most dependable SMTP server for your company. You will congratulate yourself!
Share on Twitter.