|The survey sample
included a spectrum of various people with interests in SSDs.
- 18% - said they already use SSDs
- 22% - said they were planning to use SSDs
- 46% - said they had a strong interest in SSDs
Why Do People Use
Solid State Disks?
- 10% - said they were just curious about SSDs
The #1 reason given was - Application speedup -
cited by 76% SSD buyers.
The #2 reason given was - Environment is unsuitable for hard disks
cited by 31% SSD buyers.
Budgets and Cost Expectations
had a realistic idea of how much they can expect to pay for a an SSD system.
40% of responders said they had an SSD budget above $10,000, while
21% said they had an SSD budget over $100,000.
The higher end of that
budget range is the entry level for products like the
Texas Memory Systems. But it's clear that some users have multiple applications
for SSDs. $100K can also buy a handful of
main application environment for SSDs
Historically SSDs came from
the military - and 19% said their main application was embedded (industrial or
military tool / appliance). But the high growth in the market which many vendors
cited in 2004 - is also due to the increasing use of SSDs in civilian commercial
- 14.9% - server - heterogeneous (mixed)
- 8.5% - server - mostly Windows
- 6.4% - server - mostly Linux
- 4.3% - server - mostly IBM
- 4.3% - server - mostly Sun
It's clear that there's also a market
opportunity for power desktop users in analytics and research and development
type applications. 36.2% said their environment for SSDs was a single user PC or
- 2.1% - server - mostly Apple
User Experience with SSDs - How did they feel?
who already used SSDs were asked to rate their experience of the benefits
delivered compared to expectations.
- 65% - said it greatly exceeded expectations (I advocate others to
- 6% - said SSD experience exceeded expectations (I'd use SSDs again)
- 6% - said SSD experience met expectations (my realistic goals were met)
- 12% - said regarding their SSD experience they were mildly disappointed
(some good results but not as advertised)
It's unusual for high value computer products to get such
enthusiastic responses from users. This suggests that user evangelism could be
better employed by vendors to recommend the advantages of these products to
other prospective users.
- Nobody (zero percent) was strongly disappointed. That suggest that SSDs are
not being mis sold into applications which are unsuitable. (Unlike many server
SSD Market Inhibiters and Disinhibiters
Why isn't the SSD market
already a multi billion dollar market? That's a question that we've been
asking for the past several years. Vendors have been very interested to learn
what they need to change in the way they market their products to enable this
technology to reach its true technical potential. This was one of the most
important parts of the survey and we devoted a lot of time and effort to it.
posed the question - What would make it easier for you to buy SSD technology and
remove doubts and risks which currently act as roadblocks?
some of the replies.
- 52% - want performance guarantees. (...Later:- April 2005 -
citing this survey as the cause -
Texas Memory Systems
offered the world's first performance related guarantees for SSD products.)
- 44% - try before you buy.
- 32% - buying from a big name company like IBM, HP, Dell and Sun or
their VARs. This being ranked only #3 was a surprise and suggested that some
of the channel strategies that SSD oems are pursuing may not be the best use of
their resources. (...Later:- in January 2008 -
EMC became the first "big
name" oem to launch an enterprise SSD product - saying it would use
STEC in its Symmetrix
DMX-4 high-end networked storage systems.)
- 18% - want upgrade scalability roadmaps. That makes sense - because
the investment in an SSD system can be the same order of magnitude as the
investment in the attached server. (...Later:- in April 2008
the Flash SSD Performance
Roadmap - which predicts industry SSD performance upto 2012.)
only 6% cited lower prices - as a way to get their SSD business. This suggests
that either the cost benefit arguments of SSDs compared to the technological
(non SSD) alternatives (such as more servers) are already well understood - or
that that the alternatives are just not feasible.
- 18% - supplying SSD installed with server.
The fact that over
half the responders cited performance guarantees as a gating factor to buying
SSDs suggest that users have seen far too many unmet promises about performance
in other aspects of their IT experience. They buy more storage - because they
need it. They buy more servers - because that's the accepted thing you do when
you need more performance. Nobody got shot for buying more servers. Users would
like to buy into the SSD acceleration idea - but vendors are going to have to do
more than publish attractive case studies and applications articles to get
their business, although 20% said this was important too.
commented that it would be useful to have independent information which reviewed
the performance of SSDs from various companies. One problem with this approach
is that the application speedup you get in practise is going to vary according
to the hardware environment and application. The pure performance of the product
doesn't tell you the whole story. For example product A may be 50% faster than
product B - but if the I/O is constrained by other elements in your own system -
then there may not be a worthwhile difference in the application performance
between the two products even though there's a big difference in price. Using a
car analogy - the theoretical difference between the top speed capability of a
Porsche and a Jeep is academic if you're driving all the time in city traffic
jams, or on speed restricted roads. Then again your choices will be different if
you're driving off-road. If your application needs a rugged SSD then low
weight, low acoustic noise (for use in submarines - suggested by one responder)
and low power consumption can be just as important as speed.
information such as Storage Performance Council (SPC) benchmarks or magazine
reviews while providing an independent guide to SSDs, won't answer these kinds
of questions. There's no substitute for benchmarking in your system (try before
you buy) - or vendor guarantees - if the vendor feels confident enough that they
know enough about your type of setup.
User Needs by Interface Type
is a bewildering time of choices in the high performance storage market -
because new interface types now overlap in performance with the long
established legacy duo of Fibre-channel
and parallel SCSI.
Therefore SSD vendors must either offer more connection options to users - or
risk losing customers as the market for SSDs fragments among many competing high
speed standards. As SSDs are a very technical buy it was clear from the survey
results that users are already making plans which will use emerging new
standards and render obsolete many older product lines.
We posed the
question - Which interfaces best suit your current or future SSD requirements?
- 53% - Serial ATA
(SATA) (...Later:- no SSDS were available with SATA interfaces when this survey
was conducted. A year later - the market had responded - and in Q405 - there
were 4 SSD manufacturers offering SATA.)
...Later:- as predicted by
this survey - the most popular interface type based on STORAGEsearch.com's
reader pageviews in the summer of 2006 was SATA. Sample size over
20,000 SSD content readers)
- 25% - Bus (includes PCI,
PMC, SBus, VME etc)
these results - which show users future needs - are compared to the actual
current market availability of products as listed in
the Solid State
Disks Buyers Guide it will be observed that a considerable gap
exists in the areas of SAS and SATA - which could significantly impact the
businesses of companies which are able to fill them with suitable products
- 20% - Ethernet (includes NAS,
iSCSI and IP/SAN)
SSD Packaging and Form Factors
also asked - Which SSD form factor(s) are most suitable to your requirements?
surprisingly - the #1 most popular cited - by 75% of responders - was "traditional
hard disk package (2.5", 3.5", 5.25" etc)"
came in 2nd - suiting 30% of users. This represents a smaller market by volume.
But most rackmount SSDs cost a multiple of a single SSD price - so the total
market value may be similar.
Adapter cards (PCI, cPCI, VME, PCMCIA
etc) came in 3rd - suiting 23% of users.
SSD Brand Awareness
SSD market is still at a small size (around $1 billion annual revenue) but
it's growing fast (50% per annum in 2004) and like many disruptive markets has
the potential to pick up speed and grow even faster. Measuring brand awareness
in the early stages of a disruptive market isn't a reliable pointer to which
companies will actually end up dominating the market. But measuring how brand
awareness changes over a period of years can be a useful indicator to vendors,
business partners and users.
Users were asked - When thinking about
SSD vendors - which company would you think about first to seek more
In order to minimise bias and skew in these answers -
links to the survey were randomly varied over time and position to provide equal
visibility to SSD content to advertisers and non advertisers. (Approximately one
third of all SSD vendors were advertisers at some time during the period of the
The best known SSD brand was - BiTMICRO Networks (22% of
responders). The second place was tied between M-Systems and Texas Memory
Systems (10% each.)
As a quality check on the survey we deliberately
left out some very small SSD vendors from the named list - but this didn't stop
them getting mentions in the "other" input box. This was useful to us
as a publisher too - because some of the "other" vendors suggested by
readers were companies we hadn't heard about. So we checked them out and added
them to our SSD directory and buyers guide.
Update on SSD Brand Strength
In 2007 we
started publishing a quarterly updated list of the
Top 10 Solid
State Disk Companies based on search volume. That provides a more reliable
and up to date snapshot of SSD brand awarenes.
Geographic Distribution of SSD Responders?
to the survey were asked - Where are you normally based?
The percentage in the US is about 10% below what we
estimate to be the distribution of US readers for STORAGEsearch as a whole. But
this discrepency is not thought to be significant in this content.
The Q404 STORAGEsearch.com Solid State Disk Buyer Market
Survey - has provided useful information to vendors about what they have to
do to meet buyers' needs better. If vendors take note and change the way they
market and fill the technology and service gaps which have been identified -
it's likely that the market will grow even faster in the year to come.
I'd like to thank everyone who helped design the survey, all those who
responded, our SSD advertisers and content contributors, and all the readers
whose growing interest in this fascinating subject triggered off the idea to do
the survey in the first place and made it worthwhile possible. Thanks to you
Zsolt Kerekes - Editor/Publisher [ Zsolt@STORAGEsearch.com ]
|...Later:- many SSD
oems told me they found the results from this survey invaluable.|
of the links back here include:-
in November 2005 - STORAGEsearch published a highly acclaimed
penetration model which analyzes the "user value proposition" in
the 4 main segments in which SSDs will be used in the period 2005 to 2008. This
(below) is how we announced it.|
are Most Analysts Wrong About SSDs?|
24, 2005 - STORAGEsearch.com today published a new article about the
SSD market called- "Why are Most Analysts Wrong About Solid State Disks?"
Most analysts and editors of other computer publications don't really
understand the solid state disk market. They show their ignorance and naivete by
prefacing every discussion of SSDs with a superficial analysis which compares
the cost per byte of storage between flash and hard disk drives.
the wrong answer to the wrong question. And it's far removed from why the SSD
market is racing to become a multi billion dollar market seemingly in blithe
ignorance of the cost per byte proposition.
This new article tells you
what's important to users and the main applications in which SSDs are already
being used and new applications where they will be used in the next 3 years.
In 2003 I predicted that "the SSD market is going to be a much bigger
market than the NAS market is today." I realize now that was an
understatement. So why are most of the analysts wrong? There's a simple
explanation. ...read the
article, Solid State Disks
|SSD Myths and
Legends - "write endurance"|
the fatal gene of "write endurance" built into
disks prevent their deployment in intensive server acceleration
applications - such as RAID
was certainly true as little as a few years ago.
What's the risk with
This article looks at the current generation of
products and calculates how much (or how little) you should be worried.
|RAM based SSDs have been
used alongside RAID for years - but
flash SSDs are
physically smaller and have bigger capacity (upto 412G in 2.5", 832G in
3.5") and are lower cost than RAM-SSDs and could actually be configured
in standard RAID boxes. F-SSDs aren't as fast as RAM based products but a single
flash SSD can deliver 20,000 IOPs - which when scaled up in an array - starts to
solid state disks|
|Z's Laws - Predicting
Future Flash SSD Performance|
|A few months ago a
reader asked me a very good question. |
"Is there an industry
roadmap for future flash
That prompted other questions like...
- How fast are flash SSDs going to be in 2009?, 2010? or 2012?
- What are the technology factors which relate to flash SSD throughput and
There wasn't a simple answer I could give at the time. Clues lay
scattered all across this web site
and in my many one on one discussions with readers about the market...
- How close will flash SSDs get to
RAM SSD performance?
||But I agreed there should be
a single place on the web where these answers could be found. |
Law. That gives you the wrong answer, and this article explains why. ...read the article
|Are MLC SSDs Safe
in Enterprise Apps?|
| This is a follow up
article to the popular
SSD Myths and
Legends which, a year earlier demolished the myth that flash memory
wear-out (a comfort blanket beloved by many
RAM SSD makers)
precluded the use of flash in heavy duty datacenters.|
article looks at the risks posed by MLC Nand Flash SSDs which have recently
hatched from their breeeding ground as chip modules in cellphones and morphed
hard disk form
|| It starts down a familiar
lane but an unexpected technology twist takes you to a startling new world
|How We Conducted the SSD
In the summer of 2004 STORAGEsearch, the leading publication
in the SSD market, announced in its
news pages that
it would run the SSD Survey. We contacted every known SSD manufacturer
(including some which were then in stealth mode) by email to ask them to
suggest questions which would be likely to help them understand the needs of
users better. After designing a draft version of the survey questionnaire, we
went back to vendors to ask them to comment on the questions in more details. We
incorporated some of the suggested changes and froze the design of the survey.
is an archived
version of the online questionaire.
The web based survey ran in
the 4th quarter of 2004 on STORAGEsearch.com. Respondants got to the survey by
seeing a link or image based message on our SSD web pages. We varied the
exposure during the period to avoid annoying readers. Approximately 0.5% of
readers who saw the image link clicked to reach the survey - which was signalled
as taking a few minutes to complete. In fact the 10 question survey may have
taken longer. Over half the people who saw the size of the survey dropped out
and are not included in the results.
There are 2 factors which indicate the high quality and reliability of
(1) - the percentages to the key answers didn't change much in month 2
and 3 of the survey. That shows a consistent view in the market.
(2) - some of the responses given in the "others" box are
also valid and give confidence in the sample of responders.
We used cookies to deter readers from completing the survey more
than once, and also monitored the responses daily to guard against unusual
atcivity. We chose the sample period of a calendar quarter to make sure that
this would not reduce sampling errors. Our web logs had previously indicated
that this type of high value product has a degree of seasonal interest related
to user budgets.
In total 50 completed questionnaires made up the raw
data for the survey.
...Later:- in later years we published SSD
buyer preference market reports using
search-volume based stats.
They were based on the biggest SSD focus group in the world - over 1
million readers / year.
|how to make
"SSD reliability" believable|
|Editor:- July 28, 2010 - StorageSearch.com published a
new article -
the cultivation and
nurturing of "reliability" in a 2.5" SSD brand.|
Reliability is an
important factor in many applications which use
you trust an SSD brand just because it claims to be reliable?
seen in recent years - in the rush for the
SSD market bubble -
many design teams which previously had little or no experience of SSDs were
tasked with designing such products - and the result has been successive waves
of flaky SSDs and
SSDs whose specifications
couldn't be relied on to remain stable and in many products quickly
degraded in customer sites.
||As part of an education
series for SSD product marketers - this new case study describes how one company
- which didn't have the conventional background to start off with - managed to
equate their brand of SSD with reliability in the minds of designers in the
embedded systems market. ...read the article|