|
|
new to SSD? - new report from Forward
Insights
Editor:- January 31, 2012 - Forward Insights
has recently published a new report -
SSD Technology and
Applications: A Primer (88 pages $1,499).
Author Gregory Wong says - "It's
an ideal guide for novices interested in acquiring a basic understanding of SSD
technology and applications as well as a handy reference for more experienced
professionals."
Editor's comments:- Among other things
(see contents
pdf for more details) Greg says the report also provides an overview of the
competitive landscape for SSDs.
See also:- SSD
market analysts.
Nimbus does that "no spof SSD" thing
Editor:-
January 31, 2012 - Nimbus
Data Systems today
announced
its entry into the
high availability
enterprise SSD market with the uveiling of the company's -
E-Class systems -
which are 2U rackmount SSDs with 10TB
eMLC per U of
usable capacity and no single point of failure. Unified interface
support includes 10GbE,
FC, and
Infiniband.
Nimbus
software (which supports upto 0.5
petabytes in a
single SSD file system) automatically detects controller and path failures,
providing non-disruptive failover. The E-Class also supports online software
updates and online capacity expansion. It has
RAID protection and
hot-swappable flash, power, and cooling modules. Pricing starts at $150K approx
for a 10TB dual configuration system.
Editor's comments:-
Nimbus seemed incredulous at my immediate reaction to the preliminary info they
sent me. I said I knew of competing shipping SSDs which were denser, faster
and offered more HA features too. But that's not to understate the value of
what the company does. Instead of being impressed by a bunch of me-too
technical metricals I was rather more impressed to learn that Nimbus is still
profitable. More about that later.
SSD link appears on EMC.com
Editor:- January 30,
2012 - EMC has
launched an
SSD
link (effectively still "under construction") on its
main home page.
I had no
foreknowledge of this when I wrote last week about EMC's stealth mode SSD
business - "When you start seeing a permament flash SSD link on EMC's
home page
- you'll know that the company is taking SSD more seriously."
I
was asked recently if I thought that EMC would turn out to be (in the
Christensen sense) -
like
Kodak (in photography) and
Wang
(in word processing) - yet another example of a company which - while being a
leader in one type of technology - would fail to make a successful transition
through to being leader in a a disruptive technology which would replace it.
Digital electrronics replacing optical film - for Kodak, software and PCs
replacing wordprocessors - in the case of Wang, and SSDs replacing HDD arrays -
in the case of EMC.
The thought had occurred to me too - and it has
given solace to many SSD company founders - who compete with EMC - because it
has behaved like a company which is "clueless" from the SSD
leadership perspective - despite having had the benefit of many intensive
evaluations of leading enterprise SSDs.
Nevertheless - some of the
same SSD companies which have enjoyed the ease with which they have grown their
SSD petabyte market
share at the expense of the SSD clueless EMC - would possibly change their
tune if EMC would deign to acquire their companies or oem their products.
I
told my inquirer that I hadn't quite written EMC off in the SSD market -
because as long as they retained enough loot from their rotating storage empire
they still had plenty
of time to pursue a strategy which was a blend of home grown with
oemed and acquired
SSD technologies.
The name chosen for its new SSD launch is the same
as that of another well established
SSD brand -
Lightning - from SanDisk.
But that's just a confusing coincidence - because EMC's new PCIe SSD product
will instead be based on LSI's
WarpDrive.
HA enterprise SSD arrays
Editor:- January 26, 2012 -
due to the growing number of oems in the high availability rackmount SSD market
StorageSearch.com today
published a new directory focusing on
HA enterprise SSD
arrays.
In my past 20 years of publishing enterprise buyers guides
- I've developed an instrinct for judging when the market is ready for a new
focused directory. Sometimes I've been too early - but with the momentum in the
enterprise SSD market and the number of HA SSD vendors already dipping into
double digits - I think this is exactly the right time for such a new
directory.
will rental break through the indecision barrier for SSD ASAPs?
Editor:-
January 26, 2012 - One of the business development obstacles facing
enterprise SSD ASAP
/ caching vendors in the past few years has been that users have mostly
thought of them as being HDD array accelerators.
And even if a user is
interested right now - and even if they are happy with their try before you buy
results - they often hold off making a purchase - because they think (after
reading web sites like this one) that one day they'll be ripping out their
rotating RAID systems and
replacing them with SSDs - so it might be silly to buy an SSD cache appliance
right now - if it only speeds up HDDs.
Now in reality - most users
won't replace their entire HDD storage as quickly as they might like to think -
and ASAPs do have a permanent role in the pure SSD datacenter too. Some
vendors' marketing materials talk about that - while others are still
harping on
about hard disks and the "superiority" of SSD - even when their
technology roadmap works just as well for SSD.
Seemingly breaking
through the user indecision barrier - Dataram today
published a customer story about their
"no
long term commitment" - Acceleration on Demand - leasing program. It
sounds like a good idea - but I don't know the exact terms and conditions
involved.
Fusion-io's revenue nearly trebles, but...
Editor:- January 24, 2012 - Fusion-io
today
announced
that revenue for its 2nd quarter ended December 31, 2011 was $84
million - which is 2.7x its revenue in the year ago period.
Editor's
comments:- like many other SSD companies nowadays FIO lost money in the
quarter and you can see the gory details by clicking on the links above and
going to their web site.
I'm not a financial guy - but I have
written an article below in which I share my thoughts about why loss making
SSD companies like Fusion-io are still warming (rather than cooling) SSD
interest in the
VC investor climate.
What follows includes pure speculation on my part which may be entirely wrong.
like the weather - VCs have changing climates
Because
of their pedigree the founders of Intel had
easy
relatively acces to venture capital but they aimed to be profitable as soon
as possible because the business culture of startups was very different in the
late 1960s and early 70s than it is today. Intel's early success meant
that some VCs were more receptive to the computer / semiconductor industry. But
it still wasn't easy for the company which created the first mass market for
PCs -
Apple
- when they went shopping for money in 1976 . And although the VC tech funding
climate warmed up in the early 1980s - it was still tough on founders as you
can see in this video about
Compaq. But things
were getting easier - and by the mid 1980s anyone with a good product,
strong partner(s) and a business plan could get a couple of rounds of VC
funding (including yours
truly). It was getting almost too easy - so some VCs got picky in the mid
to late 1980s with the JAW generation -
Just
Another Workstation.
dotcom lemmings
In the mid to
late 1990s in the
dotcom bubble (pdf)
I saw investors seemingly lose their sense of perspective and ability to reason
as they over funded too many nutty dotcom businesses which had no prospect
whatsoever of being profitable - based on the wild notion that growth was worth
getting if you got enough eyeballs on your site.
I was publishing a
buyers guide to dot-in-dotcom
compatible servers at the time - and I couldn't understand why people
couldn't see that many of the startups which bought these servers and never even
unpacked them - were flaky. Didn't people realize that while it was good to get
visitors to your web site - it wasn't so clever if the true cost was $100
per click. I - on my part - didn't appreciate that in a bubble it's making money
along the way which is the driving force for most investors - not actually
arriving at the end. Following the dotcom bust at the end of the 1990s those VCs
who still had money avoided most new digital investments - except Google -
like the plague.
How about the SSD market?
2 years
ago I said that we were starting an
SSD Bubble.
Nevertheless tangible benefits are being delivered to users along the way
and at the end of the rainbow will be a
huge market for SSDs.
So there are bubble elements - but some chewy goodness too. How does this relate
to the many companies in the enterprise SSD market today who are growing
revenue - but not necessarily profitable - like Fusion-io?
My view
about Fusion-io's rolling losses is that part of this is due to the
continuing investment (in technology, sales and marketing) which any similar
company has to make in a fast changing. fast growing tech market - but another
factor in its profit equation may the high proportion of its business which goes
to a small number of big customers. It's just a fact of life that when
storage companies sell to server oems and super users they have to sell at a
lower price than if they're selling to other types of customers - because there
are competiutors out there who will also buy this business opportunity. But even
in the dearth / absence of profit in such deals- the high sales volumes which
result - speed up positive outcomes in other factors which can be healthy
for future business development.
It is to be hoped that at some point in the future - as the innovation curve
flattens - and the technology creator's brand strengthens and the product
becomes a sticky standard supported by compatible 3rd party partners - the
margins in the product itself and in the channel mix may change for the better.
(Licensing deals
too are another possibility for extracting more profit from high volume oem
customers.) There are no guarantees in any competive market but that's my way
of trying to make long term sense of what's going on in some hot spots in the
SSD market today.
Intel buys InfiniBand line from QLogic
Editor:-
January 24, 2012 - Intel
yesterday
announced
an agreement to acquire the
InfiniBand
40Gbps (pdf) related product lines, IP and business assets of QLogic.
Editor's
comments:- if you're not familiar with
InfiniBand - it was
originally proposed in 2000 as a standard for remote CPU R/W with small packet
sizes and ultra low latency to support arrays of CPUs over many cards and racks.
In the early days - InfiniBand evangelists and some storage analysts believed
the standard would go into the commercial server mainstream.
Instead
what happened was that fatter multi-core CPU chips, and faster GbE wiped out
the volume market need for IB technology - because they could do the same job
cheaper and incrementally for smaller clusters of CPUs. So the IB market
nowadays is mainly a niche market for scientific research and high performance
computing.
Some of the
fastest SSD benchmarks
have been recorded in IB environments. And at one time (before 2008) I thought
that IB might be a significant and natural upward path for high performance
SSDs. However, PCIe SSD
systems also support remote array connections - so IB's role remains that of
occupying the narrow turf of clustering hundreds to thousands more CPUs than
Intel or others can pack into a single chip.
Another way to think about
it is this. You can't have viable HPC without SSD. But you can have a healthy
SSD market where HPC is a small niche.
There's no doubt that SSDs are
an enabling technology which make it realistic for CPU designers to think
about what they could do with hundreds of cores on a single chip and over 1,000
cores on a single server card. I discussed that blue sky concept with
processor designers nearly 10 years ago. But does the mainstream market need
such servers?
In the
SSD
data driven factories of the future - the answer is yes. But that could be
another 5 years in the future - because there are still closely related
standards to firm up - such as Hybrid
Memory Cubes. And
storage history
shows that new standards
take years to get into the market. In the meantime - if you're not in the HPC
market - but still need very fast CPU performance - keep an eye on what the
leading PCIe SSD makers do - and you won't go far wrong.
TCS ships 200GB fast erase MIL-STD-810 2.5" SSD
Editor:-
January 24, 2012 - TCS today
announced shipments of a rugged 200GB 2.5" SLC SSD which has has been
verified by outside labs to meet MIL-STD-810 requirements for shock, vibration,
temperature range, temperature shock, humidity and altitude.
The new
Galatea
SSD has 40K IOPS performance, includes 128-bit AES
encryption and can
fast erase the full
drive in less than 15 seconds.
"Few solid-state drives combine the
quality, data capacity and ruggedization features of Galatea," said Michael
Bristol, senior VP and GM of TCS' Government Solutions Group. "It
is ideal for a wide range of extreme industrial and defense applications,
including oil and gas exploration, avionics and data logging in a variety of
air, land and sea vehicles. Galatea combines superior access latency and power
consumption performance with long-term reliability."
Editor's comments:- I hadn't heard of TCS before in the SSD
market - and I feel uncomfortable when I see a significant new SSD product pop
out from seemingly nowhere. But then I recognized one of the legacy products
names - Triton and sure enough TCS is the new identity for
Trident Space &
Defense - which was acquired a year ago.
I googled "Galatea"
- and I'd like to think it was named after one of the
Harry Potter characters
- who taught defence against the dark arts.
Later today:- -
Charlie Cassidy who
is Director of the Advanced Products Group at TCS contacted me to say - "I
thought I would let you in on the "secret" of the Galatea name. No
Harry Potter involved, we didn't even realize that connection. Our SSDs (Triton,
Proteus, Galatea) are named after themoons
of Neptune - paying homage to the Trident heritage."
Violin video re visibility advantages of home grown controllers
Editor:-
January 23, 2012 - I commented recently that the
top 10 SSD companies in
Q4 2011 all had one thing in common (apart from the fact they make SSDs) -
they all had their own proprietary
SSD controller
architecture which they could use to optimize products for some application
markets (even if some of them also used other controllers too).
In
a
recent
video - Violin's,
CTO Software Jonathan Goldick
talks about the benefits they get from having their own controller.
I
like it because it also echoes themes I discussed last year in my
big versus small
SSD architecture article - and also because it's short - less than 250
seconds. ...watch
Violin's SSD video | |
| . |
|
|
| |
| .... |
 |
|
... |
|
| "Since the early
1970s there have been 3 revolutionary disruptive influences in the electronics
and computing markets:- the microprocessor, the commercialization of the
internet and the transition of digital storage to SSDs." |
| ......from the article -
Perspectives - on
the SSD market | | |
|
..... |
|
1.0" SSDs
1.8" SSDs
2.5" SSDs
3.5" SSDs
19" rack SSDs
Analysts - SSD market
Animal brands in the
SSD market
Articles
and blogs - re SSD
. |
|
|
|
. |
|
 |
|
. |
|
| How big was the
thinking in this SSD's design? |
Does size really does matter in SSD
design?
By that I mean how big was the mental map? - not how many
inches wide is the SSD.
The novel and the short story both have their
place in literature and the pages look exactly the same. But you know from
experience which works best in different situations and why.
When
it comes to SSDs - Big versus Small SSD architecture - is something which was
in the designer's mind. Even if they didn't think about it that way at the time.
|
 |
For designers, integrators,
end users and investors alike - understanding what follows from these simple
choices predicts a lot of important consequences. ...read the article | | | |
|
. |
|
 | |
