30th quarterly edition - based on search metrics
in Q3 2014
by
Zsolt Kerekes,
editor - StorageSearch.com - December 1, 2014
The Top
SSD Companies - researched and analyzed by StorageSearch.com - has
established a unique reputation within the
SSD industry for
detecting and predicting significant new business and technology transitions in
the SSD market.
I know from what readers tell me that each new edition
is anticipated and read by many founders of SSD companies, their biggest
institutional investors and customers too. It's one of the most
popular articles
on this site - and has been seen by millions of my readers - and referenced
by millions of others in indirect attributions from the companies named in these
articles.
And from a personal point of view - as I look back and
think about all the many SSD related articles I've written - I have to
admit that starting the Top SSD Companies List back in 2007 was probably the
best time to do it.
When the series started in 2007 - there were 55
SSD companies. I knew them all intimately and nearly all were private
companies.
Before then - the market was too small for people to care
about such a list. But the launch of the Top SSD Companies series in 2007 -
as is now clear from the years which followed - marked the beginning of a
new growth phase point for the market which was on the verge of making
no-turning back transitions in enterprise SSD flash controller architecture.
This would lead to
flash replacing
DRAM as the definitive enterprise SSD memory type - a move which would lower
cost and make SSDs more attractive to more users. (For a contemporary view of
those early heady days see
2007 - Year of
SSD Revolutions.)
The Top SSD Companies series has also provided
early market alerts for all the significant new technology and business
ideas which have changed in the market - and also been a dependable predictor
of which companies would benefit or suffer the most from discernible trends in
the market.
For more about the past successes of this list,
highlights from, and links to the 29 past quarterly editions - go to
the series overview page.
Here's the list of the Top SSD Companies ranked by analyzing the
search volume of hundreds of thousands of SSD readers on StorageSearch.com in the 3rd
quarter of 2014. |
.... |
Top SSD Companies - ©
StorageSearch.com
based on SSD search volume in the 3rd
Quarter 2014
For more details about each company click on the compay
name. |
Up 17 places!
Entered the Top SSD
Companies list - 2013
Q3
Diablo's 1st time at #1 is significant for many
reasons - coming as it does after the 5 year occupation of this position by
Fusion-io.
Reader
interest in Diablo in this quarter (by the SSD readers of StorageSearch.com)
was 33% higher than the #2 ranked company -
SanDisk (despite
SanDisk having acquired Fusion-io in this period).
Diablo entered the
Top SSD Companies Lists a year ago - in
Q3 2013 - so you
my ask - what made such a stunning difference to the level of interest in this
company and its technology platform in Q3 2014?
The overwhelming factor
driving the surge of sustained interest in Diablo in this qaurter was
curiosity following the unveiling of Diablo's next generation (2015)
architecture for its memory channel storage (DDR4 DIMM attached flash SSD
arrays) which the company said would have 48nS latency for data in its cache
pipelines - a stunning advance on the already significant 3µS in 1st
generation products.
As we've seen in earlier phases in SSD history
any technology which promises to bring great performance gains to
application servers - attracts a lot of reader interest.
At the
present time we're seeing a lot of new solutions being introduced to enable big
memory fabrics inside servers and enabling connections between servers.
Even
before Diablo's announcements in Q3 2014 - the company had raised important
questions about aspects of server based SSD archietcture such as:-
- is there room to introduce a new latency tier within servers?
- is there an appetite in the software market for new low latency / big
persistent memory platforms? - such as would enable new applications?
Another
factor driving interest in Diablo's technology - although much smaller in scale
- was the ongoing saga of the company's patents entanglements with a company
called Netlist (which
designs an entirely different type of product - which is a
flash backed DRAM DIMM).
Netlist
had asked the courts to restrain sales of Diablo's technology (which is the key
interface technology inside SanDisk's UlltraDIMMs). We may see that question
answered one way or the other soon after publication of this article.
See
also:- memory
channel SSDs, flash
DIMMS, MCS
versus PCIe SSDs
Up
4 places.
Best previous position - #1 in
2007 Q3.
Entered
the Top SSD Companies list -
2007 Q2 (1st
edition).
If you had asked me to predict what SanDisk's
position would be in this list - before the quarter began - I would have said I
expected SanDisk to acquire most of the
benefit
from having acquired Fusion-io - which was the previous long running #1 SSD
company.
That's partly what happened.
SanDisk attracted
3x as much interest from our readers as the jointly ranked #3 SSD
companies which are listed below.
But even though SanDisk has been
the main flash partner of Diablo - the Diablo effect - or was it the surge
realization that memory channel SSDs are really different to PCIe SSDs (and
different to hybrid DIMMs) - but in ways which are different - and not simply
interchangeable? All these factors led to the surprising result that in this
quarter SanDisk didn't simply inherit the #1 slot in this list from
Fusion-io.
See also:-
SSD services,
SSD software ,
PCIe SSDs
3 (tied) - IBMdown
1 place
Best previous position - #2 -
2014 Q1 and
2014 Q2
Entered
the Top SSD Companies list (TMS an IBM Company) -
2012 Q4
There
was no significant change in IBM's enterprise SSD offerings in this period.
Having said that - IBM wasn't entirely silent on the subject of enterprise
SSDs. In
June 2014 - IBM
announced that an IDC
report had identified IBM as the #1 company (ranked by revenue in 2013) for
rackmount SSDs with 25% market share.
Another Diablo effect?
An
interesting observation - given the prominence of Diablo in this article -
is that in November 2014 Alex
Yost - who had been VP Strategy and Alliances in IBM's x86 server business -
joined Diablo
Technologies as President.
IBM had been Diablo's biggest
customer - and had worked on refining MCS architecture for years. Yost cited
the outstanding customer benefits he'd seen at IBM from Diablo's technology as
a factor in his move.
Up 7 places from previous quarter.
Entered the Top SSD Companies
list - 2010 Q3
This
is the highest ever rank for Seagate in the 7½ years of this landmark SSD
market series.
It shows that once a company has realized it is
hostage to the
fortunes of the SSD market and decides to do something dramatic about it -
then astonishing changes can happen.
In this quarter having
completed acquisition of the SSD product lines from
LSI - there was only
one direction which Seagate could go. And that was up.
Along the way
Seagate wasted no time in telling the world what it was going to do with this
treasure chest of controller and PCIe SSD assets - launching straight into
announcements about new products and a new customer relationship with Baidu.
Same as before.
Best previous position - #4 -
2012 Q2 and
2012 Q3
Entered
the Top SSD Companies list -
2010 Q2
In
this quarter OCZ previewed the Z-Drive 6000 - its first product aimed at the
2.5" enterprise
PCIe SSD market.
For SSD customers in the central European
market - OCZ also launched a new German language website.
Down 2 places.
Entered the Top SSD Companies list (as HGST) -
2013 Q3
Related
1st entry in the Top SSD Companies list (as Western Digital Solid State
Storage) - 2009 Q2
In
this quarter HGST announced a new improved version of the
high availability
clustering capability previously available in the PCIe SSD product line
acquired last year from Virident.
Quaintly at odds with its intrinsically modern PCIe nature, however,
these HA features are implemented by
InfiniBand or
ethernet rather than PCIe
fabric.
Up
2 places from previous quarter.
Entered the Top SSD Companies list -
2014 Q1
In
this quarter A3CUBE unveiled details of its software platform and said that it
was on the verge of its first US customer shipment.
In a
conversation with the founders of A3CUBE I learned that in addition to the
many possibilities of extending legacy server apps beyond the constraints of a
single server box - which the company had already written about - they didn't
think it would be too difficult for systems integrators to also mix and match
their extended PCIe fabric with newly emerging persistent memory based APIs -
such as SanDisk's ZetaScale - or Diablo's MCS platform.
This
demonstrates that some of the new solutions which we're seeing in the enterprise
SSD market already may have the ability to scale and create new capabilities
which go way beyond what any one of them was originally planned to do by
themselves at the time they were conceived.
Up 12 places.
Entered the Top SSD Companies list -
Q4 2013
In
2014 - 2 companies have stood out for me with respect to their ability to
invent genuinely new business models in the enterprise SSD market based on
their unique analysis of customer data and sifting through the same raw
customer material which has already been well raked over and ignored by
competitors.
They are Tegile and SanDisk.
When I mentioned
this to Rob Commins,
VP Marketing at Tegile recently - he said the 2 companies have something else
in common - which I hadn't mentioned. SanDisk is an investor in Tegile.
See
also:- Exiting
the Astrological Age of Enterprise SSD Pricing,
SSD services
Down
2 places from previous quarter.
Best previous position - #2 in
2013 Q1,
2013 Q2 and
2013 Q3
Entered
the Top SSD Companies list -
2007 Q3
Before
Violin became a publicly owned company I observed that the company was involved
in too many product segments within the SSD market.
In 2014 Violin's
solution to the growing
fragmentation
and specialization within the enterprise SSD array market has mostly been
to introduce new product variations (differentiated by software) at a faster
rate than ever before.
Has it worked?
Judging by the reaction
of the market as filtered by
revenue
(23% lower in the quarter ended October 31, 2014 than in the year ago period)
- the answer appears to be not yet.
Related to that sales performance
- in December 2014 - Violin
announced
the appointment of enterprise storage sales veteran Jeff Nollette
as worldwide VP Channel Sales.
But is sales the problem for Violin?
Or is it just a symptom of complex marketing issues?
The problem (as
I see it) is that the company's energetic marketing communications output has
been engaged in the vain pursuit of multiple message wars which it can
never win because almost everything it says - sounds like exactly the
same words which other competitors have been saying too.
Violin has
some genuine core SSD strengths and some genuine technology disadvantages.
Instead of pretending it can be great all kinds of enterprise deployments -
simply by leveraging its most recently announced software compatibility - it
would be better for the company in 2015 to realize its limitations - and do
less things the best - rather than do more things - in ways which are
interesting but not compellingly competitive.
Down 2 places from previous quarter.
Best previous position - #4 in
2013 Q1
Entered
the Top SSD Companies list -
2012 Q2
Skyera's
great advantage is that because it understands how to manage cheap modern flash
memory in arrays better than any other systems competitor - it has the potential
to create very competitive high capacity flash boxes.
But Skyera's
self confessed disadvantage - told to me by its founder - is that when the
company was founded it didn't understand the enterprise market.
The
idea appears to have been to solve the difficult flash technology integration
problems first and then add the simpler "enterprise" must-haves later.
As
a result - its initial products - didn't have the best mix of features -
despite its low aspirations in the "enterprise" features list.
Skyera
tweaked the enterprise feature with an evolutionary model launched in October
2014. When I was talking to Skyera's new CEO about the changing enterprise
market - I said - it looks as if Skyera is grooming itself to join an elite old
gentlemen's club - at a time when it's not even certain if the current members
- or the club will survive long.
As I write this - the clear
marketing model for Skyera to avoid is Violin's apparent model of simply
adding more and more software to the same underlying flash array to meet some
idealized notion of what a legacy storage box should look like.
Skyera told me it was already talking to customers about new roles for
its SSD boxes which would be technically unfeasible for other array
competitors to have conversations about (no matter how much software they had
- my comment) because fitting in the physical space budget is sometimes what
makes the application viable.
...Later:- Well as things
turned out - soon after publication - that Skyera did manage to avoid following
Violin's business development model - as it got acquired in mid December 2014
and went into the enterprise SSD melting pot of
HGST.
Currently
the only scale at which Skyera's technology works is at the box level.
Will
we see it being reapplied into new HGST PCIe SSDs? - It would take a very long
time to integrate the technologies in this way. And I doubt if it would be
worthwhile.
Instead I think it's more realistic to think of the 1U
rackmount SSD - as an alternative and convenient standard form factor for
embedded enterprise SSD applications.
Up 4 places.
In this quarter
Avago Technologies completed it
acquisition of PLX.
The core technologies involved here are:- PCIe
switching chips and the emergence of PCIe as an enterprise fabric.
I
discussed PLX's fabric solution -
a 1U PCIe fabric
switch - in an article earlier this year. And I've updated the article with
more details which have emrged since.
If the PCIe fabric idea is
successful - then whoever makes the box matters less than who makes the chips.
And
PCIe SSDs are getting everywhere - with
2.5" and
M.2 form factors already
having joined the traditional server slot sizes.
So in that respect -
whatever assumptions you make about the adoption of PCIe fabric in high end
servers (compared to traditional connectivity methods) PLX / Avago can
look forward to significant increase in PCIe chip volumes coming from the SSD
market.
Down 1 place from previous quarter.
You can buy
SSD boxes built from standard 2.5" SSDs from hundreds of companies.
The
challenge for Pure is to try and convince enough enterprise users that if
they're going to buy this kind of safe sounding boring product anyway - they
should buy it from a company which does this as the main thing it does -
instead of from a vendor who sells flash arrays as a sideline because its
older storage products are no longer selling so well.
The fact that
Pure maintains a good ranking in this list of Top SSD Companies (when many
of its competitors have lower positions or have never even been in the list)
is a sign that when markets get big and confusing - a simple focused message can
be as important as technology.
Same as before.
Virtium didn't announce any significant new products in
this quarter.
Throughout 2014 I've been talking to leading industrial
SSD companies - such as Virtium - about the problems posed by market growth,
fragmentation and the difficulties of using the same flash technologies which
work in more benign applications.
From a business point of view
choosing the viable reality of excellence in selected niches above the
unfeasible goal of having the best technology roadmap for all applications - has
become a priority in the
industrial SSD
market.
I discussed the industrial market as #11 in my article -
what
changed in SSD year 2014?
Up
11 places.
2 of the topics which Foremay regularly talks about at trade
shows are:-
In
this quarter Foremay updated its long running outputs on these themes with a
paper
(pdf) presented at the Flash
Memory Summit.
In November
2014 - Foremay announced it was accepting orders for 8TB 2.5" SSDs
optimized for the
military SSD market.
Down
3 places from previous quarter.
Some SSD brands leave longer lasting
impressions than others.
So too the best way to understand Cisco's
SSD offering - for many people - still begins with reading about WhipTail - who
along with its Invicta SSD product line - Cisco acquired in October 2013.
Up 3 places.
Kaminario's guaranteed effective (virtual) capacity concept was 1 of 3 new
enterprise SSD cost models which I later grouped together in my article -
Exiting
the Astrological Age of Enterprise SSD Pricing
In
December 2014 - Kaminario
announced
it had closed an oversubscribed $53 million financing round, bringing total
raised capital to $128 million.
See also:-
VCs and SSDs,
HA SSDs
Down
3 places from previous quarter.
Nimbus didn't say anything noteworthy
in this quarter.
See also:-
rackmount SSDs
Down
1 place.
Usually when talking about Micron in an SSD context you'd be
thinking about segments like
PCIe SSDs,
hybrid DIMMs,
consumer SSDs, and
flash memory itself.
In
November 2014 (remember that was after the closing period for metrics in this
Q3 list) Micron announced it was serious about the auto market too.
Down
3 places.
See also:-
military SSDs
1st time
appearance in these lists.
In this quarter PLDA anounced the
availability of a new SSD
controller aimed at the PCIe 3.0 SSD market.
PLDA's IP has been engineered for both ASIC/SoC and FPGA
implementations.
Same as before.
Among other things - in September 2014 - Samsung
announced it had started mass producing 3.2 TB NVMe PCIe SSDs (HHHL) based on
its 3D flash memory technology, for use in enterprise systems.
Same
as before.
Up 1 place
Reentry
to the list.
what about the companies which dropped out of the list?
2
very significant SSD companies dropped out of this list simply because they
had been acquired.
EMC dropped out of the list -
for reasons you'd have to ask our readers about. But EMC didn't drop far. It was
#26.
Other companies just below the Top SSD List threshold also
included:- Innodisk
#27, Maxta #28,
Marvell #29 and
Memblaze #30.
Note:- Rankings above are based on analyzing the search activity of
over 280,000 SSD readers on StorageSearch.com in the 3 months period ending
September 30, 2014.
The rankings above are based on multiple types of search and
browsing activity - and after filtering out unreliable and fake searches
coming from robots and spam intended to subvert rankings for criminal or
malicious reasons. Our rankings and tie breakers also use data from some other
SSD related internet sources from time to time.
You can read more
about the significance and track record of this methodology in predicting and
observing SSD market intentions in
earlier editions of this
article. | |
... |
|
... |
. |
|
| |
..... |
|
.. |
|
... |
SSD news SSD history VCs in SSDs market research the SSD
Bookmarks SSD
news archive - July 2014 SSD news archive -
August 2014 SSD
news archive - September 2014 the Top SSD Companies -
series overview 10 things
which changed in SSD year 2014 4 big things which
changed in SSD year 2015 about the publisher - 22
years guiding the enterprise the previous edition of
the Top SSD Companies - Q2 2014 the next edition of the
Top SSD Companies List - Q4 2014 |
... |
A new scale of ambition -
enabled by a creative and competitive SSD ecosystem which has the
confidence to move mountains
In
2014
I think we've been witnessing the birth of a renaissance in SSD inspired
enterprise architecture - on a scale of ambition we haven't seen since the
Year of SSD
Revolutions in 2007.
Back then it was about the idea that - on the
SAN - enterprise flash
architecture could take on the role of
RAM SSDs.
Meanwhile (also in 2007) - talking about a new idea for SSDs
inside the server - the first enterprise customers were being shown by a
company called Fusion
Multisystems what could be done with
software centric
PCIe SSDs.
In
2014 / 2015 - the assumption (originally foretold by
PLX) is - PCIe SSDs are
everywhere - so what next?
The answer seems to be - adding depth,
breadth, globalized-viewpoint efficiency and what I call - "continuum
SSD architecture."
The details come in many shapes and sizes.
At least one more
tier of faster SSD latency inside servers.
Maybe more
tiering within
DRAM too.
New hardware and software fabrics which will have the
same effect on how you come to view a single server - as
RAID did on the
limitations of a single hard drive.
So anything you could do before in
a single server - can now be done on tens, hundreds or thousands or servers -
without having to throw away your legacy applications.
From the
cloud we've learned
the value and flexibility of being able to incrementally augment and scale
resources. But we also know that sometimes the cloud isn't always there when we
need it.
SSD vendors are bringing cloud-like manageability and
micro-tiering into small spaces (tiny clusters of servers or storage) which
you can afford to own. So you don't have to trust any assets which aren't in
your own direct control.
New
SSD based software
will demolish many limits you thought were set in concrete - such as the
size of server memory and the restrictions of how you can redeploy and migrate
data which is siloed in different legacy systems.
Those old limits
were set - by performance and software based on ancient rules carved out over
20 years ago.
The new SSD revolution - is building on the confidence of
an ecosystem - which starts with this idea...
We made some things
faster and cheaper and easier - with SSDs - picking at the little easy
problems we could put in a single business plan.
Now we know how to do
this stuff - and SSDs are everywhere.
So let's change everything and
make it all much better.
what
changed in SSD Year 2014? | |
... |
|
... |
There are many segments for
enterprise flash arrays which aren't listed or even hinted at in standard models
of the enterprise market.
Many of these missing market segments don't even have names.
Hey - that means SSD-world is like a map of the US before Lewis and
Clark.
If you're a VC
should this make you anxious or happy? |
Decloaking
hidden segments in the enterprise for rackmount SSDs | | |
...
... |
|
...
... |
|
... |
|
... |
"I think
SanDisk's ZetaScale could be one of the most significant SSD software products
launched in 2014 - because of the freedom it will give big memory customers
about how they navigate their tactical choices of low latency flash SSD
hardware." |
Zsolt Kerekes,
editor - StorageSearch.com
- SSD news (July
8, 2014) | | |
...
... |
Tegile's
IntelliCare and SanDisk's STAR programs are both interesting examples of a new
marketing differentiation trend in the SSD market - where some companies are
making it easier to do business with them - by offering services which make it
easier for customers to buy their SSDs.
|
services which sell SSDs | | |
...
...
...
... |
|
...
...
...
...
...
...
...
...
...
...
...
... |
Skyera is putting a lot of
effort into joining a market which resembles an old fashioned English
gentlemen's club.
But we know the founding members of that club are
so old they will die soon anyway.
Is it worth it? |
SSD news story -
skyHawk FS - October 29, 2014 | | |
...
... |
|
...
... |
"I think the context
in which to view this is as the embodiment of a new wisdom in the industry -
that to succeed in the enterprise SSD market today - and to achieve the ultimate
efficiencies at the manufacturing level - vendors have to think like systems
companies." |
Zsolt Kerekes,
editor - StorageSearch.com re news that WDC
had acquired Skyera (December 15, 2014) | | |
...
...
...
... |
Today - if you're in a
big company in a traditional market - and hoping to do something equally big in
the SSD market - then $1 billion may not be enough - but $5 billion may be too
much. |
VCs & SSDs | | |
...
... |
|
...
... |
Something in the market has
changed.
I noticed this week that the topic of HA/FT SSDs has risen to be 1
of the top 10 topics that you've been looking at this month. Which means it's
mainstream.
|
high availability
enterprise SSD news (October 13, 2014 ) | | |
... |
|
...
... |
|
...
... |
...
... |
|
... |
|
... |
|
... |
who are the 10 plus $1B eSSD companies?by
Zsolt Kerekes,
editor - February 16, 2015
You're reading one of my SSD list series of
articles now - but soon it may be joined by another.
A new special
interest list of SSD companies - those whose annual revenue in the enterprise
SSD market is running at a $1 billion or more run rate.
When I started
thinking about it - I realized I could - without much thought - populate such a
list most of the way up to 10 companies - from piecing together things I've read
in presentations and news stories. Which indicates to me that with a bit of
research - and help from readers - we should easily get past 10 by the end of
2015.
So I will publish a preliminary list of who these companies are
and which products are the most important ones driving their revenue in Q2
2015.
If you think your company should in there - just
contact me and
let me know the details. | | | |