An SSD conversation with Tegile by
Zsolt Kerekes,
editor - June 23, 2014 |
|
By instinct I'm more interested in pure
solid state storage products than hybrid drives or hybrid arrays - so 20 months
ago (in October 2012) when Rob
Commins, VP of Marketing
Tegile Systems
contacted me to ask about the
Top 20 SSD Companies List
- and whether I kept an eye on
hybrid array vendors
- what I literally said was this.
"In recent years my view has
been that hybrid auto acceleration systems (SSD ASAPs) do have a
long term place in pure enterprise solid state storage (as described in my
paper - an introduction
to enterprise SSD silos). But some products which have been optimized
around the temporary (5-6 year) market tactical role of working with HDDs are
better able to make to transition to pure solid state hybrids than others. We
(by which I mean my readers and I) are more interested in products and
architectures which look like they will survive and adapt into the pure SSD
world. We have much less interest in systems which include HDDs."
If
that sounds like it's kind of defensive and dismissive it is.
That
kind of conversation is even shorter today because nowadays I'm talking to
over 600 real SSD companies - and I can't spare much time for the hundreds of
other companies which aren't even committed enough about solid state storage to
put SSDs into every slot in their box.
Anyway - going back to Rob
Commins and Tegile...
Although we kept in touch after that first
encounter - and I spoke to Rob and some other people connected with Tegile
from time to time - the time budget which I allocated to learning about Tegile
and my editorial coverage of his company - was fairly skimpy for the next
15 months or so - until - surprise! surprise! - the
4th quarter of 2013
- when enough of my readers had picked up on my minimal coverage of Tegile
up to that point to propel the company into entered the
Top 20 SSD Companies
List.
At which point - despite my reservations about hybrid arrays
which use COTS SSDs - I had to sit up and take notice.
So in that
edition of the list - published in January 2014 - I observed that...
"Tegile
Systems (ranked #19 in Q4 2013) is the highest ranked vendor in this edition of
the Top SSD Companies List whose primary business is hybrid SSD arrays - rather
than SSD drives or SSD systems."
And so I started paying a bit
more attention to Tegile. Because when enough of my readers show a
disproportionately high interest in something - that must mean there's
something to it. (Even if I have missed the early signals myself.)
How
easy is it to recognize the portents of those rare companies which
emerge from stealth or crowd but have the potential for SSD stardom?
In the case of some new unknown companies (like the early
SandForce, or
Violin or
Fusion-io or
Skyera) it was
obvious to me from the very first conversations I had with those companies
that that what they were doing would resonate with a large proportion of
people in the SSD market. The only thing which took me by surprise in those
earlier cases was the speed of your response.
But it can also happen
that a company - like Tegile - which I initially think of as doing stuff which
is a bit samey - and which I don't think too much about at first contact -
plants roots in enough of your minds - by the things they do - which results
in high search volumes which leads me to look again at what they're doing - and
look for a possible cause or narrative which I may have missed the first time
around.
That's one of my safety nets.
And in the case of some
companies - it works so well that I start watching out for an important news
story - hours or days before I get it in my email.
Anyway - going back
to Tegile - once they hit the Top SSD Companies List - I started paying more
attention.
And when that status sustained and improved slightly in
the following quarter
too - despite the company not doing anything which to me seemed out of the
ordinary - I thought I'd better talk again to Rob Commins and educate myself a
bit more about his company.
Which is how we got to this article
you're reading now.
I warned Rob in advance - I'm not interested
in any new products you may be launching. And especially not hybrids. I can
read about those on your web site.
Just stick to telling tell me more
about your technology and the way you do business.
Because I know
there's something you're doing which I'm missing - but which has been making
waves with my readers.
Anyway - that's what led to a recent
conversation in which I learned some new (to me) things about Tegile's
technology and Tegile's business model - which are different to what I expected.
Tegile's
hybrid array technology - the different thing
To preface this I
have to mention that when you've seen as many SSD caching, tiering, compression
and dedupe products as I have - you realize that all of them are permutations of
the same handful of architectural possibilities. So I'm not going to go into
any of the details here.
The key differences I learned - compared to
the options I expected were:-
- Their system has a lot of RAM
(compared to some other
skinnier flash
arrays).
- So a lot of the thrashing around activity - which you expect to see in
this kind of system - which is related to the metadata - occurs in the RAM.
The result is better performance (than if it was all stuck together) and lower
R/W demands on the flash array (which gives better reliability).
- Users can vary the proportion of flash to HDD they buy in their systems -
but they can also vary the ratio they deploy - in the configurations they have
already purchased too.
Tegile has a slider switch graphic as the
metaphor for this concept. (click to expand the
image)
- The same system software runs in Tegile's hybrids as in their pure flash
arrays.
I realized when talking to Rob that their storage array
behaves - in effect - like a micro-cloud in a box which is
under the users' control.
Rob said that cloud utility model is
something which Tegile leverages in the way it does business. And that's what
took me on to the business model. |
 |
product overview -
IntelliFlash Storage Arrays from Tegile |
Tegile's business model - the different thing
Rob
told me that some time ago Tegile realized that although customers preferred to
buy the cheapest system they think they need (with a higher ratio of HDD to
SSD) some of these same customers - if given the freedom to switch on a
higher ratio of flash (if it was already in the array) would experiment with
using more of the flash enhanced virtual features - and having learned to like
the results they got - would settle at a higher enabled flash ratio than maybe
they would have initially purchased.
That's why the company came up
with a concept the called
Agility Pricing.
With
Agility Pricing - selected customers can opt to pay for their storage arrays
on a utility model - in which they pay for the virtual storage they use rather
than the raw storage which is installed in the box which Tegile supplies them.
The
financial decoupling is handled by a traditional type of finance company.
But
in effect - this scheme means that Tegile can leverage its cleverness at
magnifying virtual storage from deploying its flash and software utilization IP
- in a business model which is transparent to the customer and which can
generate revenue in proportion to the efficiency of Tegile's array
architecture.
From the viewpoint of resellers - who sign up such
customers - the reseller gets rewarded in a similar way to if they sold the
system.
Depending on the customer and their applications - this could
mean users would be paying under 5 cents per virtual GB per month.
In
effect - Tegile puts virtual flash capacity and software in at one end and
gets a guaranteed revenue stream out at the other.
I said I
thought that was a brilliant business idea.
That kind of flash to
dollars
pipe used
to be the exclusive domain of the
SSD speed kings
- but now - while sticking resolutely in the SSD slow lane - Tegile has
invented a new formula which converts SSD software trickiness and marketing
savvy into a business proposition which anyone can understand.
I
could also see how Tegile's initial immersion in the hybrid array market gave
them advantages compared to pure play SSD array vendors - when it comes to
understanding the needs of customers in the low end and mid range of the
enterprise market.
By offering a continuum of price and performance in
the same product line Tegile gets exposure and business from customers from a
wider spectrum of markets and use cases than vendors of traditional monolithic
SSD arrays.
Traditional rackmount SSD vendors (which Tegile now
competes with as Tegile moves up market to pure flash arrays) have mostly
started at the high end of the market with very fast systems and are now
trying to move down the cost and performance curve - which they can
accomplish from the technology point of view.
But SSD vendors moving
down the budget food chain lack a true market understanding of user
issues in these entry level and mid range projects. The business credibility
problems they face are a bit like when the minicomputer companies in the early
1980s tried to market PCs - such as the
DEC Rainbow 100 -
which was admirable when you saw it at a trade show - but utterly unaffordable
because it included too many useless features which were put there to
demonstrate how clever its engineers were - rather than to meet genuine PC
customer needs.
the power of try after you buy
The
flexibility of Tegile's technology (in particular the ability to change the
characteristics of the usable capacity by experimenting with ratios of HDD : SSD
compression etc) means that users can learn more about their own needs and find
their own best comfort zone for operating their storage in a relatively risk
free and low cost way.
And - just as important - from my point of
view - Tegile's technology - which enables them to magnify the usable virtual
capacity - and to monetize it in an acceptable way - works in a pure SSD
storage context too.
summary
In this conversation I
learned more about Tegile's internal cloud like technology. But I was even
more impressed by a new (to me) business model which Tegile has been testing
on selected customers.
I can see now why the early
investors liked this
company even though it took me longer to get it - because I have a rule of thumb
filter which predisposes me to discount the relevance of companies don't
have their own flash
controller IP and whose storage arrays are mostly populated by
hard drives.
Here
are some related links:-
- SSD ASAPs
(Auto-tuning SSD Accelerated Pools of storage)
| | |
.... |
Hmm... it looks like you're seriously
interested in SSDs. So please bookmark our home page -
StorageSearch.com - and come back
again soon. |
. |
About the publisher - 22
years guiding the enterprise market |
. |
| |
.... |
 |
.. |
 |
.. |
|
..
.. |
If you've seen or read -
The Hobbit - then you'll be familiar with the concept of the riddle game.
Something similar is playing out now in the enterprise flash array market. |
playing the enterprise
SSD box riddle game | | |
..
.. |
There have always been
storage heirarchies in the data storage market - the old rotating types of
storage - such as hard disks, optical drives, HDD RAID arrays and tape evolved
with software and processors over many decades and so too did ways of dealing
with them.
The new economics of SSD storage took the CPU designers and OS
software developers by surprise.
Instead of weaving in SSD support into computer architecture over a
10 year period - based on an incremental technology roadmap - the SSD market has
gate-crashed the server party - and the SSD roadmaps are changing too fast for
the old style computer vendors to keep up.
|
the New Business Case
for SSD ASAPs | | |
..
.. |
|
..
.. |
"Bottlenecks in the
pure SSD datacenter will be much more serious than in the HDD world - because
responding slowly will be equivalent to transaction failure." |
will SSDs end my
bottlenecks? | | |
..
.. |
|
..
.. |
Who are the top SSD
companies?... the companies which you absolutely have to look at if you've got
any new projects involving SSDs? |
the Top SSD Companies | | |
..
.. |
|
..
.. |
| |