The inventors of Shared Internal Storage (SIS),
Paris-headquartered Seanodes is changing network storage technology. Seanodes'
SIS platform Exanodes radically alters the economics and possibilities in
data storage and application processing. Seanodes has earned multiple awards
from industry analysts and media for its outstanding technology that virtualizes
storage assets to convert unused internal disks and Direct Attached Storage
(DAS) into a shared storage array. Founded in 2002, Seanodes is headed by
storage industry veterans from two continents and backed by a number of private
equity firms. More information can be found at www.seanodes.com or by calling
- editor's comments:- in June 2009 -
of tests using entry level SSDs
virtual SAN software.
In an ESX environment of 8 servers with 1 SSD
drive per server, IOmeter benchmark results showed 36,000 IOPS (random read 4K)
for a system with an overall cost under $20K (including the cost of SSDs and
Exanodes VM Edition).
"'Traditional arrays have been designed to work efficiently with
spinning disks and can't give the promise of SSDs in terms of performance and
scalability for example," said Frank Gana, Business Development Director at
Seanodes. "This limits the usage and markets and as a consequence most
people use them as Direct Attached Storage with all the usual known problems
that come with DAS. Thanks to Exanodes and its innovative design we can
aggregate and use SSDs efficiently, opening new markets and applications to this
Seanodes says it's trying to fix the problem of
aggregating and sharing multiple low capacity, low cost SSDs between servers
without requiring special tuning skills. But I have to say the quoted IOPS don't
sound impressive to me compared to the
fastest SSDs. So
why wouldn't you use less servers and a better SSD instead?
many other competing solutions in the
rackmount SSD and
PCIe SSD market - I
suspect that Seanode's solution may only provide an economic price point for a
tiny fraction of possible applications - or none at all. There isn't enough
data in the press release to be sure.
|"In 1994 -
a couple of SSDs appeared in my buyers guide (which was aimed at the
SPARC compatible server market). One of these - the T8000 - was an
80MB, 10MB/s SSD accelerator with internal real-time data compression on a
single slot SBus card, made by Colorado based CERAM. Units in multiple
slots could be chained to appear as a single SSD with upto 960MB capacity.
Performance was 2,000 IOPs."|
36 year rise of the SSD market|
|the Problem with
Write IOPS |
|the "play it again Sam"
Flash SSD "random write IOPS" are now similar
to "read IOPS" in many of the
why are they such a poor predictor of application performance?
why are some users still buying
RAM SSDs which cost
more than SLC and significantly more than MLC? - even when the IOPS specs look
||This article tells you
why the specs got faster - but the applications didn't. And why competing SSDs
with apparently identical benchmark results can perform completely
is published by