|
|
leading the way to the
new storage frontier |
..... | |
| |
....... |
more like this? -
later and
earlier in the StorageSearch.com news archive |
.... |
end of the line for OCZ's
investors
Editor:- November 29, 2013 - Just before the holiday
OCZ
announced
its working credit lines had been terminated - forcing the company to
file for bankrupty.
OCZ also said it has received an offer from Toshiba to acquire
substantially all of its assets in a bankruptcy proceeding provided various
conditions are satisfied - which preserve the value of OCZ's business
including the retention of employees.
Editor's comments:-
whether it ends up being Toshiba - or another company - which gets the
value and benefits of leveraging OCZ's SSD assets will depend on the outcome of
legally controlled auction processes.
2 weeks ago I spoke to OCZ's CEO
- Ralph Schmitt, and OCZ's CMO - Alex Mei and was able to
confirm some key assumptions and guesses about OCZ's technology and the
demographics of their enterprise customers. For details see the article on
the right.
Under new ownership some of these developments -
if properly resourced and leveraged - could have a bigger market impact
than would have been possible within the constraints of OCZ's financial
straightjacket.
See also:-
PCIe SSDs,
SSD software,
VCs and SSDs
what
are others saying?
Jim Handy,
Objective
Analysis - "If the Toshiba acquisition proceeds.... Toshiba has
not had as strong of a presence in the SSD market as the company would like, and
OCZ would give it strong technology and a respected name in the retail market.
Meanwhile, Toshiba would give OCZ something that the fledgling firm has so far
been unable to attain: a steady source of NAND supply at competitive prices.
(OCZ has had a difficult history of having to pay price premiums during NAND
shortages owing the company's consistently-poor credit ratings.) For both OCZ
and Toshiba the synergies of a takeover are good. If the deal falls together
Toshiba may become a real powerhouse in the SSD business."
SSD reliability is tools deep
Editor:- November 26,
2013 - Although we often talk about
SSD controllers and
flash memory chips as
being "hardware" - none of these devices would exist if it weren't
for a rich software ecosystem of design automation and verfification tools
which have evolved to enable semiconductor system design
in the past 30
years or so.
So the
reliability of SSDs
depends not just on the interplay of physics, architecture, electronic
factors and firmware - but just as importantly - reliability or more
accurately the dependability and determinability of any SSD design begins with
the verification and integration of all the design abstractions - in
software.
A salutary reminder of the importance of this came today in
an
announcement
from Hyperstone
about its use of such tools from Synopsys
in its design process for controllers aimed at the
industrial and
medical markets.
"Our new
S8
SD 3.0 and eMMC 4.4 industrial flash memory card controller is designed to
deliver the highest level of reliability and data retention when using
innovative sub-20nm MLC flash technologies" said Axel Mehnert, VP of
Marketing at Hyperstone. "Our mission is enabling future NAND flash and
advanced technologies to be fit for use, especially within industrial,
ruggedized applications. This requires the robust verification and analysis
provided to us uniquely by Synopsys' VCS functional verification solution. No
other technology had the capacity, performance, advanced verification features
and reliability to verify such complex designs."
Editor's
comments:- in reality it's never as simple as - we used these tools -
turned the handles - and the design popped out. May work for simple devices like
CPUs or DRAM - but not for real complicated stuff like SSDs. Imagination, true
grit, magic spells and serendipity are essential elements in SSD alchemy too.
Violin's 1st quarterly report
Editor:- November 25,
2013 - Now that Violin
is a public company - it has to publiicly report revenue etc - and in its
first such
report
for the quarter ended October 31 - revenue was $28 million - which the company
says was 37% more than a year ago.
"Enterprise data center
storage is in the early stages of a major transformation to an Enterprise Memory
based infrastructure, and Violin is at the forefront of accelerating this
transformation," said Don Basile,
CEO of Violin Memory.
Editor's comments:- one of the
self-awareness repositioning adjustments which Violin has still yet to make -
is that compared to the time when the company entered the SSD market 6 years ago - when there were
only 59 companies in the whole SSD market - and only
2 of the top 10
companies at that time (including Violin) were marketing
rackmount SSDs -
we're now in a market where the number of SSD companies is 10x bigger. I
could easily give you a list of over 100 vendors today which market
rackmount SSDs or
hybrids and - in the most recent quarter -
7 of the top 10 SSD
companies were marketing rackmount SSDs within their product line.
All
of which means - there is a much bigger market which Violin could get - if it
were not for the pesky inconvenience of all those other competitors.
So
- when the company says that the enterprise market is still at the "early
stages" of adopting SSD systems - in one way I'd agree with that - insofar
as the upside potential for this whole market in revenue is more than 10x more
than it is today - but on the other hand - while Violin has got good
brand
recognition and has established a track record in some segments of the
enterprise SSD market - for example in
high availability
and fast SSDs - I
don't rate it as a strong runner in the race for some of the
new
technology developments which are emerging in the rackmount market.
Nevertheless
- the segments in which Violin is already arguably a leader - are big enough.
I often say to CEOs in enterprise SSD companies that an important
business strategy for surviving to play in the bigger SSD market of the future
- is to get better at what they do now - and to focus on just one or two
strategic product types which they can execute well and be the #1 or #2 -
rather than making the mistake of enlarging their product lines - and being an
"also compete" company which is never quite good enough to get the
business - because it operates in too many diverse application segments in
which it can't possibly be the best solution.
what are others saying
about VMEM?
NetApp has shipped 59PB of SSDs in past 3 years
Editor:-
November 19, 2013 - Among other things - Network Appliance
today
disclosed
it has has shipped over 59 petabytes of flash storage in the past 3 years.
Editor's
comments:- What NetApp actually said was "over 60PB to date".
My calculation goes like this...
The company shipped 1PB in
its first year in the SSD market - which ended in the 3rd quarter of
2010.
So
it's shipped approximately 60PB in 3 years. Probably more than 1/2 of that will
have been in the past year.
How does that compare with others?
It doesn't sound like a lot in the context of today's market.
According
to a (quirkily pro-HDD)
blog
by Toshiba - the
analyst data which
they have
aggregated
projects that 8,000 PB of enterprise flash SSDs will ship in 2014.
I think the likely figure (PB of enterprise flash installed in
systems) will be much higher than even that - because Toshiba's data probably
doesn't take adequately into account the ability of some systems vendors to ship
enterprise grade SSD racks using consumer grade flash chips (rather than using
enterprise SSD drives) due to technologies like
adaptive R/W
- and the increasing appetite for enterprise SSDs. See also:-
petabyte SSD shipment
milestones.
SSD companies named in Fast 500 list
Editor:-
November 19, 2013 - I found 2 SSD companies in the Technology
Fast 500 List (pdf) published today by Deloitte LLP.
#4 - Fusion-io
- 832x growth
#445 - OCZ - 1.6x growth
The
ranking is based on revenue growth in the period from from
2008 to
2012 - for
companies above a threshold level and which satisfy some other criteria.
More
info about the list here (pdf)
Editor's comments:- I read
through the list fast - so if I missed an SSD company let me know.
2013 - wasn't
such a good year for revenue growth for the companies named above - so it's
unlikely they would do so well in next year's edition.
LSI integrates "SSD market on a chip"
Editor:-
November 18, 2013 - LSI
today
launched
its 3rd generation SandForce SSD controller family - the SF3700 which - based
around a single chip design - spans a wide spectrum of SSD market
applications (from consumer to enterprise) - includes native jumper-selectable
SATA or gen 2 PCIe interfaces - and incorporates adaptive R/W DSP ECC
management.
Editor's comments:- The SF3700 (now sampling) is
the most ambitious design of a single chip
SSD controller in
SSD market
history.
Its 14 core design integrates many impressive design and
architectural features including:-
- the ability to efficiently configure as either
a small
architecture or big architecture SSD controller.
The SF3700
design can be configured with as little as 3 flash chips in entry level
consumer SSDs - or as many as 129 chips when maximally configured in a 9 channel
enterprise design which can recover from the complete failure of a memory chip
as well as partial failures in other memory chips in the array.
- dynamically adjusted power islands within the chip - enable a single
silicon design to support both the low requirements of deep sleep mode in SATA
notebooks as well as the performance requirements of entry level PCIe SSDs.
I
recently spoke to Kent
Smith at LSI about this new product.
Our conversations about
SandForce SSD controllers go back more than 4 years - so we skipped a lot of
stuff.
One of the first things I said to Kent - was - I've been
nagging you for years and asking - when are you going to do a native PCIe SSD
controller?- and for nearly 2 years it's been clear that another big hole in
LSI's SSD IP bag has been
adaptive R/W
- and now you've finally done both at the same time in a single product.
I
was also really impressed by the quality of LSI's
briefing
document on the LSI SandForce SF3700 (pdf) - which explains just about
everything you need to know. So I asked Kent - why does he need to waste time
talking to editor's like me? - why doesn't LSI just publish the document on the
web and let it speak for itself?
I said a lot of publications will
simply copy some of your pictures without attribution - and I think readers
would find it valuable seeing them too - but I think it would be fairer to the
work you've done if I could just make the whole document available - so there
was no doubt who had done the hard work of communicating what the design was all
about.
He agreed to that - and you can click on the link above to see
the original info which I got from LSI.
Some other things I learned
from this conversations were:-
- The SF3700 is a completely new design. - It leverages all the flash
related design concepts related to endurance and array level fault management
which have been proven in earlier designs and extends them too.
For
example RAISE has been enhanced so that for high-end configurations it can
protect against a full memory chip failure as well as multiple block faults -
whereas entry level SSDs which need some RAID like features but can't afford an
extra memory chip can use fractional RAISE.
- One of the reference designs which LSI offers for this controller is for an
M.2 form factor - which is
goiing to be the game changing SSD for the
consumer market
next year. The card design is the same whether the SSD is being used as a
SATA or PCIe SSD. A single jumper sets the configuration at assembly time.
LSI's 2.5" reference design will also make it easier for oems to
produce products for enterprise arrays in the
2.5" PCIe SSD market. Overall
I think the SF3700 is a very ambitious and outstanding SSD controller design -
which will elevate LSI's reputation within the SSD industry.
For the
past year or so I had been wondering if the glory days of LSI's SSD controller
technology lay mostly in the past. But I can now understand why it took them
so long to integrate this new design - which is almost at the integration level
of "SSD market in a chip".
A design which integrates so many
architectural features which are optimized for so many markets wouldn't have
been feasible for a small SSD start up.
Below you can see one of the
many pictures I spoke about in LSI's paper. If you click on it you'll see the
whole thing. |
|
PS - I almost forgot to mention
one funny marketing thing I
learned.
Kent told me that they used to call the SandForce products "SSD
processors" but then found that didn't show up too well in web searches -
because people were looking for "SSD controllers" instead. So LSI has
changed its parlance and is now calling them "SSD controllers" too.
As
I said above - Kent draws some great pictures which illustrate the
functional blocks within SSDs.
He also writes a lot of SSD
blogs too. So I was relieved to hear that he gets paid on an SSD
marketer's pay grade rather than that of a writer. That means SSD editors
and bloggers don't need to worry that he'll be tempted to come and replace us.
Everspin's MRAM caches Buffalo's industrial SSD
Editor:-
November 18, 2013 - 18
months ago it was reported that Buffalo Technology was
designing MRAM as the cache in a "soon to be shipping" new SATA 3 SSD
aimed at the industrial
market. Bringing that up to date - the identity of the MRAM supplier is now
known to be Everspin
Technologies - which today
said more
about this application.
""As an early adopter of ST-MRAM,
Buffalo Memory is taking a bold step to continue as an innovator in the SSD
market," said Phil LoPresti,
president and CEO of Everspin. "Spin-Torque MRAM technology will give
Buffalo Memory a strong differentiator in the market for high-performance
industrial SSDs."
Editor's comments:- Everspin's
64Mb DDR3
ST-MRAM (pdf) obviates the need for supercaps in
sudden power
fail situations while providing very fast write latency (compared to
flash). Other past applications for these chips include flight recorders and
write logs in RAID systems.
Although
skinny flash SSD
controllers can also solve some of the same problems - and also don't need
supercaps - they do need more complex data management within the
controllers to work dependably.
Phison controller inside Chromebook
Editor:- November
14, 2013 - Phison
Electronics today
announced
that its PS3109
controller is being used in an SSD inside the
C720
Chromebook launched recently by Acer.
Maxta joins the elite set of enterprise contenders who are
vying to own the next generation SSD-centric platform
Editor:-
November 13, 2013 - This week Maxta completed its
staged emergence from stealth mode and
launched
its first product - the Maxta Storage
Platform - a hypervisor-agnostic software platform for repurposing arrays
of standard servers (populated with cheap standard
SATA SSDs and
hard drives) into
scalable enterprise class apps servers in which the global CPU and storage
assets become available as an easily managed meta resource with optimized
performance, cost and resilience.
Editor's comments:- I spoke
last week to Yoram
Novick about this new product, his company and what customers have been
doing with it.
Before you dip into my bullet points below - here's a
header note of orientation.
We've all seen new companies launching
SSD software and pitching for the enterprise with products which are little
more than spruced up versions of "hello SSD world!"
Then a
year later - some essential compatibility features get added, and later still
some degree of better or worse
high availability.
It didn't used to matter much if everything wasn't in place at the start - or
if these new companies didn't have sustainable business plans - because there
was an appetite for acquiring
them.
From my perspective I'd say that many companies have
regarded the launch of their SSD software is simply an invitation to attract
users who could provide the market knowledge they needed to flesh out the
feature set.
In these important respects Maxta is different because:-
prior to this week's product launch they've already had a group of 10 or so
advanced customers in different industries who have been using the product and
also the enterprise features - like manage-ability, scalability, resilience and
data integrity are already in the product today.
Maxta's technology and
business architects have done enterprise storage software before - as you can
see from their linkedin
bios. Yoram told me that he and
Amar Rao (Maxta's VP of
Business Development ) used to compete with each other in earlier storage
startups and the companies which had acquired them.
So it soon became
clear to me in the details I saw and asked about (not all of which are listed
here) that a lot of careful planning and up front thinking and problem solving
has already guided the "launch".
Here's some of what I
learned.
- market scope
MxSP is the software glue for enabling easily managed
SSD enhanced storage
pools in VM environments which scale from the ROBO upto
cloud infrastructure.
The
base level configuration which provides HA features starts as low as 3 nodes.
This is attractive for enterprises with remote offices because it's a small
footprint. But it's also attractive from a running cost point of view too -
Yoram said because of the special low price point for associated software.
Maxta
has a customer who started with these 3 node configurations for remote offices
but liked them so much that their bigger arrays are now built mostly from
arrays of 3 too.
- the problem it solves
The evolution of enterprise CPU and storage
resources have followed different tracks in the past decade - leaving users in
the position today where it's easy and economic to deploy more CPUs but
relatively awkward, expensive or error prone to map these CPU resources into
virtual storage which scales with the same ease and which takes advantage of
the low cost and high performance of commodity enterprise SSDs.
- the storage pool
Maxta's architecture aggregates the SSDs and HDDs
in the server pool into a single globally accessible, fault tolerant SSD
accelerated virtual storage pool.
Within Maxta's software - all the
SSDs are collected together as 1 super SSD resource and another big resource
is created from the HDDs.
Internally Maxta's software knows that SATA
SSDs and SATA HDDs have different personalities for example:-
- HDDs have low cost per unit of capacity but slow random read latency
- SSDs have fast random read, and fast sequential write
Not
every node in the array has to have an SSD or HDD inside but it's not sensible
to have a system which doesn't have any SSDs at all.
- fault tolerance, data integrity, VM snapshots, cloning etc
Yes
they're all in the product now.
- software? - it's a virtual world view
Everything about MxSP is
virtual. And it doesn't require new management tools. The operational aspects
will clarify in customer case studies and white papers.
- Maxta's business plan
I told Yoram how disillusioned I had become
about the sustainability and viability of new storage software companies -
given my experience of having tracked over 1,000 storage companies and
terminating the list of
gone-away and acquired companies in a single decade at the 500 company
level. (That's before I started the gone-away SSD companies list BTW - which is
well on its way to 100.)
Jaundiced by that experience it seems to me
that over 95% of storage software startups don't have much of a clue about how
to translate their IP assets into any sustainable business value and are mostly
founded at the outset with the fervent desire that before the VC and IPO
money run out - they will get acquired. So I asked him if Maxta would be any
different to that?
Yoram told me some of what Maxta has been doing in
laying the foundations for growing the business to become a significant storage
platform (in his words) a significant software company like Microsoft or
Veritas.
I won't say more here because this is too long already -
despite having not even mentioned most of the notes I made during our
conversation.
Looking back on this nearly a week later (and having seen
some of their documents before) I'm left with the impression that maybe indeed
Yoram is right and his company could become not only one of the rare storage
software companies which are sustainable as a business. But going further than
that - maybe too it has the makings of a company which could be one of the
five to ten companies which will dominate the SSD software platform market of
the future.
Who are the other contenders?
I've given you
lists before - but this list is evolving because 4 of the 10 companies were
still in stealth mode last time I did that.
If you're interested in
the SSD enhanced storage platform idea (and who wouldn't be) then another good
place to look is the list of competitors which I've compiled in
Maxta's profile page.
Rick and Dave's new stealth mode enterprise SSD company -
Primary Data - gets $50 million initial funding
Editor:-
November 11, 2013 - you may be asking yourself - how does a company founded
less than 6 months ago - and which doesn't say anything much on its web site -
manage to attract $50 million in its first external funding round?
The
answer is given by one of these investors, Ping Li at
Accel Partners - who - in the
recent
funding press release - said "Primary Data's biggest asset is its
team of recognized pioneers in the storage industry. Their technical strength
and remarkable experience in successfully bringing disruptive technologies to
market will help to solve a number of data virtualization and mobility problems."
- who's Primary Data?
- Primary Data, founded in 2013 and
with operations in Salt Lake City, Silicon Valley and Israel - says it is "developing
next generation data virtualization and mobility technologies to manage how
information is stored and shared globally."
- what are they up to? - (best guess)
SSD software which
will change the economics of
cloud servers and
storage by disruptively improving the utilization
efficiency and
performance of virtualized apps servers.
Editor's comments:-
now if you've been researching this - you'll see that I haven't actually told
you any new details about Primary Data which you couldn't already have learned
from seeing the public statements which the company has dispersed in various
places on the web.
One of the peculiarities about stealth mode
companies is they do need to talk to people to get things done - but they also
need to control the flow of information about what they reveal for various
reasons - not least of which are:-
- they don't want to alert competitors
- they might change the details and direction of what they do
Last
week I was talking to the founder of yet another stealth mode SSD company* -
whose web site says a little bit more than that of Primary Data - but I can't
say who he is yet either.
I can tell you what he told me, however,
which is this...
He had to launch his minimalist web site some time
before he would have ideally liked to - because some of his early customers
have policies which mean they can't make payments to a company which doesn't
have a web site.
Going back to Primary Data - I can tell you this.
I
confidently predict that when Primary Data does formally exit stealth mode
and launch its new products - it will join an elite band of SSD companies (Fusion-io,
SandForce,
DensBits and
Skyera) in entering
the Top SSD Companies
List within a single quarter. (If not before).
* - later note added - November 12 (Tuesday) - I can now tell you this
other conversation (about the web site thing) was with Yoram Novick,
founder of Maxta - which is coming
out of stealth mode today. There will be much more about this soon.
memory channel SSD vs PCIe SSD write latency - 3rd party
benchmarks
Editor:- November 7, 2013 - Although it's possible to
make estimates of how a new type of
SSD interface
will perform compared to what already exists - the interplay of applications
software, systems software and a new
controller design -
means that initial assumptions can easily be out by a factor of x2.
And even if you have accurately guessed what the weak points and the
strong points are in the product's
performance
symmetries - it's only when you start to see
benchmark results
that you can have more confidence in your
models of what's
happening inside.
So if you've been wondering how Diablo's
memory channel
storage compares with PCIe
SSDs -
click
here to a new whitepaper (pdf) which includes some useful data. The
application isn't important but it's the first public glimpse which goes
usefully beyond the graphs shown in the product launch documents.
The
highlight for me is - a mean write latency of about 30µS for MCS compared
to about 100µS for PCIe SSD- at a particular R/W ratio which may of
course look nothing like your own setups. |
|
When you stop and think about
it - we might see just as large a difference between
different
models of PCIe SSDs - and as we don't know the identity of the mysterious
"PCIe SSD" in this paper - and as we've only got a very limited view
of the comparison - you might say this whole paper is null data.
On the
other hand - being able to show that in one situation an MCS SSD can be 3x
faster than a hypothetical PCIe SSD - suggests that the new MCS SSDs aren't
slouches either.
My best guess back in
April 2013
was that there would be no material performance difference between an MCS SSD
and a best of breed PCIe SSD - when working with best of breed VM optimized
software APIs.
I think the differences between memory channel and PCIe
for server SSDs shouldn't be seen simply as performance differences (which may
indeed not be that significant) but as different integration approaches
which offer a different set of branching out directions depending on the
software environment and the
high availability framework.
The 2 types can even work together in the same server in different roles.
Later:- 3 years later - the next generation of DIMM based
memoryfication products from Diablo (Memory1) resulted in more clearly
differentiated benchmarks - this time making comparisons with DRAM arrays - see
the article - risk
reward ratios with big memory "flash as RAM" - for details.
InnoDisk's new FlexiArray
Editor:- November 6, 2013 -
Until today - if anyone had asked me - what do you think about InnoDisk? - I'd
have said - they're an
industrial SSD
company which has done some noteworthy permutations on
MLC flash care and
adaptive
controller IP.
I was surprised therefore to learn that they've got
aspirations in the rackmount SSD market too. Or that's what they think they're
doing. The company will demonstrate its
FlexiArray
1U and 2U racks later this month at the
SuperComputing 2013 conference.
Editor's
comments:- Personally I suspect this may be one of those embarrassing
marketing episodes when a company which knows a lot about one technology -
mistakenly believes this is enough to make it a significant participant in
another market - which it really doesn't understand at all well.
You
may think differently - but InnoDisk's
1U
SSD boxes sound underwhelming to me - and appear to bear a closer
resemblance to some naiive products I saw being offered in 2006 - rather than
the much more cunningly
engineered rackmounts of today.
It's always possible I'm wrong and
that their SSD software is more capable than it seems in their online
documents.
Component makers yearn to get into the
rackmount SSD
business - because that's where the real money is. But it's a very
sophisticated market and you need more than just a bunch of
2.5" drives in a
box and some firmware to impress anyone that you're serious.
...Later:-
- It turns out I was wrong about this product.
In 2014 -
I learned why this apparently bland system is ideal for some types of customers
- and I also learned that the firmware inside owes more to modern controller
thinking.
See:-
Decloaking
hidden segments in the enterprise for rackmount SSDs
EMC really doesn't like Pure Storage
Editor:-
November 6, 2013 - EMC
is suing Pure
Storage.
There's a good report on this in AllThingsD.com.
EMC's premise is that it can't understand how Pure could have been so
successful in picking up EMC customers in any other way than by leveraging the
detailed knowledge which former EMCers brought to Pure when they were recruited.
Pure's
CEO - Scott Dietzen
- tells the world what he thinks about EMC's move
in
a blog in which he says - "While I have no insight that would allow
me to comment on EMC's motivation, I would say in general more mature companies
risk forgetting the golden rule - they are happy to recruit great people to join
their companies from competitors (indeed they aggressively solicit such hires),
but then resort to onerous non-compete agreements and lawsuits to deter the same
employees from exercising their freedom to seek employment elsewhere..."
Editor's
comments:- I was talking to the founder of another enterprise SSD company
earlier this year about something different - but as a part of that conversation
he told me about how upset companies like EMC, HP etc really are about what
they regard as SSD upstarts causing them huge business pain.
I can't
remember the exact ratio he used - but the gist was that from the big
incumbants' point of view - whenever they see an upstart SSD company get one
million dollars of business say - that effectively loses the big company maybe
five to ten million dollars of business which they believe they would have got
if these SSD companies didn't exist.
For most of us - that big
company viewpoint is ridiculous.
Because in the same way that any
rational customer in the late 1980s was going to buy a PC instead of a
minicomputer to do their wordprocessing, or in the early 1990s - a Unix based
server from
Sun etc to do
their departmental database and email instead of a mainframe from DEC or
IBM.... customers today know they can save tons of money and do more stuff by
switching to new SSD companies for their server and storage needs.
The
main thing slowing them down switching over to the SSD highway is the
bewildering array of SSD
route maps on offer. (Not all of which have pavement and gas stops.)
Knowing
these self evident truths about technology transitions doesn't stop the pain for
the big companies like EMC who feel like they have been
coerced
into playing a game - which they don't like or
understand.
This
legal move by EMC reminds me of another pivotal moment in SSD history - in
April 2008 -
Seagate fights SSD Market Challenge - with lawyers - instead of engineers
- when Seagate
hoped to crush STEC
using the same tactics which had worked before with several inconvenient
hard drive startups.
That didn't work for Seagate in
2008. It
couldn't wish the SATA SSDs away.
In 2014 - the
rackmount SSDs
and PCIe SSDs aren't
going to disappear either.
You could argue that if EMC's proprietary
knowledge of
what do
enterprise SSD users want? was such a valuable business asset - how is it
that EMC itself didn't do a better job in the SSD market while the talent was
still on the payroll?
Is LinkedIn
going to be the next target for the SSD lawyers? - After all - that's where you
can advertise what you know.
Or how about the storage conferences and
trade shows? Are the stars from EMC going to be forced to present their papers
anonymously - so they can't be grabbed by competitors?
One the other
hand - if it wasn't for the lawyers - we wouldn't have all those great books to
read by John Grisham. I'm currently
rereading - A Time to Kill.
Coho Data gets $25 million for 2U SSD ASAPs
Editor:-
November 5, 2013 - Coho
Data (which recently emerged from stealth mode and operates within the
hybrid systems - SSD
ASAPs market with a cloud
market focus) - today
announced it has
got $25 million in Series B funding - led by new investor, Ignition Partners,
and also including existing investor, Andreessen Horowitz.
The funds
will accelerate Coho Data's R&D and go-to-market efforts as the company
prepares for general availability of its
Coho DataStream system later
this year.
Editor's comments:- What I like about Coho Data is the clarity
about what it offers:- "Build your own high-performance Amazon-style
storage for all your data."
What I'm less gung ho about, however,
is the apparent efficiency.
For example - one of the "difference"
features which Coho talks about on its website is the easy way that users can
scale to multi-petabyte installations - using its 2U building blocks which have
Intel
PCIe SSDs inside and
some hard drives.
The
example shown on this
Coho page says you can build a 190TB hybrid apps server which does
900K IOPS in 11U of rackspace for $530k list. That may sound good compared to
museum grade alternatives - but Coho's solution takes 2x to 3x
the rackspace of leading pure SSD HA based systems to do the same thing.
Now
- size isn't everything. And neither is scalability. And for many users the
ease of managing incremental growth might be compelling reasons to look at this
type of product. But most of you pay more per square foot for your floor space
than Amazon - so an "Amazon style" product proposition may not be the
best for you.
new blog by PernixData describes the intermediate states of play
for its HA clustered write acceleration SSD cache - from fault discovery
to recovery
Editor:- November 5, 2013 - In a clustered,
SSD ASAP VM
environment which supports both read and write acceleration it's essential to
know the detailed policies of any products you're considering - to see if the
consequences - on data vulnerability and performance comply with strategies
which are acceptable for your own intended uses.
In a new blog -
Fault
Tolerant Write Acceleration by Frank Denneman
Technology Evangelist at PernixData
describes in a rarely seen level of detail the various states which his
company's FVP goes through when it recognizes that a fault has occured in
either server or flash. And the blog describes the temporary consequences - such
as loss of acceleration - which occur until replacement hardware is pulled in
and configured automatically by the system software.
Stating the design
principles of this product - Frank Denneman says - "Data loss needs to be
avoided at all times, therefore the FVP platform is designed from the ground up
to provide data consistency and availability. By replicating write data to
neighboring flash devices data loss caused by host or component failure is
prevented. Due to the clustered nature of the platform FVP is capable to keep
the state between the write data on the source and replica hosts consistent and
reduce the required space to a minimum without taxing the network connection too
much." ...read
the article
SSD ASAPs - auto tiering /
caching appliances high availability
enterprise SSDs | |
. |
SSD news
SSD news archive -
2013
SSD market
history - since the market began |
. |
| |
|
|
|
..... |
..... |
scary Skyera the fastest SSDs SSD symmetries top SSD companies what
changed in SSD year 2013? the enterprise SSD
software event horizon latency loving
reasons for fading out DRAM in the virtual memory slider mix |
|
. |
|
|
|
. |
|
So - you think you
already know this leading SSD company? (OCZ)
What can I say to
change your mind?
|
Editor:- November 21, 2013 - You'd think that in this age of SSD rumors,
tweets and superbloggers - and particularly given the nature of
the SSD market
itself being - as it is - so intimately bound up with the sub-atomic
technologies of disseminating information quickly around the far corners of the
internet - like some kind of higs-boson data-quark on steroids -
gathering in the distant CPU server clan atoms while concurrently
pre-accelerating their local impact by collapsing space-time - that it
would be remarkable indeed not to know everything it's important to know
about the recent achievements of a
leading SSD company
- which has already been stared and glared at publicly for longer than seems
decent or comfortable - and which is rarely out of the news - even when the
company itself is trying to have a rest day in the SSD PR factory.
You
could say - more than enough has been written about this company already. And
it's almost certain you've engaged with them in some way - or thought about
them. You've got your views safely tucked away in the mental box which holds a
small bunch of SSD companies whose names you can remember - and about which you
can recall some interesting stories - without having to look at a web page.
If
anyone were to ask you - what do you think this company might be doing next
year? - compared to where the company thought it was heading a few years ago -
you'd probably say - it's like one of those high flying rockets where the engine
misfired or the wrong numbers were typed into the navigation system. We're
seeing a lot of those in the SSD market. - And maybe one day one of those
rumor vapors which have been misting around the company for the past few years
might come true. But you don't think they are the force they used to be - and
while their products retain a firm foothold in the market - you see that
more as a consolidation of what they did before - rather than as a springboard
to new greatness.
Like you - I have those attitudes too about a lot of
SSD companies.
Unlike you - I'm observing more companies - so the
number of seconds or minutes (each year) that I can devote to thinking about any
individual company (easily calculable) - is pretty small. Which means that
whenever I reach a decision about how to compartmentalise any particular
company - I do it fast. It's done.
Working on web time I don't have
any realistic alternative to that - "decide in haste" thing but I
neither do I hanker after "repenting at leisure" either - so to
be on one the safe side - I also periodically revisit my firmest held
assumptions from time to time to see if they're still valid.
You can
see lots of examples of this revisonism if you trawl down my past comments
about any SSD company which has been in the market for more than a couple of
years.
If they change what they do - I change my views. And the
landscape of the SSD market is changing all the time too. So sometimes the
strategy which was right for a company in one year is the wrong thing to do in
another. You get the general idea - and I'm sure it works the same for many of
you too.
Now I know that some of you - SSD leaders, analysts,
investors, users and even SSD bloggers - often spend much more time thinking
about a single SSD company or small number of companies than I could ever do.
When we talk - as we sometimes do - you often know a staggering
amount of pinpoint pixel level detail. I'm more into seeing an
impressionistic wider picture.
You see a new product.
I
see a bunch of decisions predetermined by architecture and location (time and
place) in the market.
You see this quarter's revenue, cash-flow,
operating expenses and analyst conference call.
I see this as just
one of many raindrops which will one day form an SSD ocean.
Despite
those differences - our conversations Dear Reader often end something like
this.
You update your model. I update mine. We all think we see a
little clearer than we did before.
So how can I get you to change your
mind about a company about which you already have very clear and set views?
Time
for me to get to the point.
Like many of you - I've been interested in
seeing how OCZ
has been doing under the stewardship of the crash response team under the
leadership of Ralph
Schmitt - who became the company's CEO in
October 2012
and who - almost exactly
a year ago - was
reported to have said - "we've got the train back on the track".
In
hindsight that derailment analogy - which referred to OCZ's predicament in the
first half of 2012 was apt - because even though the train can still be been
seen chugging along at a more cautious speed in about the same direction
- if you turn around and look back from whence it came - you can see a lot
of parts have been left strewn by the wayside - and it's only recently that the
train regulator has felt confident enough to stick back on the front of the
engine a new clean shiny badge of safety conformance. |
|
On the surface - a few weeks ago - it sure
looked like everything that one needed to know about OCZ was already being
reported on and being said. But I felt uneasy.
I had seen several
clues that the company had been doing much more than simply fire-fighting and
fixing a derailment - and if this had been any other SSD company then either
they (or analysts like me) would have been talking much more about some of these
matters. But those other aspects were either hidden behind the smoke of the
clearup operation or not deemed important enough by the company to talk about -
as their likely consequences would be more quarters in a future which was
already fogged by rumor and speculation.
So I thought it would be a good time to either confirm or clarify what
was happening at OCZ in a number of strategic areas. And the only way I could do
that would be to have a chat with Ralph Schmit. So I put in a request - and the
answer came back almost immediately that he looked forward to having a chat
with me too.
So this is a summary of some of the things I talked
about last week (November 14, 2013) with Ralph Schmitt
CEO and Alex Mei
CMO at OCZ.
As I hadn't spoken to Ralph or Alex before I thought I'd
start by saying that I wasn't interested in asking them to comment about any of
the acquisition rumors which have been circling around their company. I
mentioned another recent rumor - related to another SSD company - which was
ridiculous for anyone who knew the specifics of the 2 named companies. I told
Ralph I think these speculations are often started by people who have vested
positions in one of the company's shares.
I also know that what I
write can have an influence. So people try to influence me. Apart from the
obvious lobbying for attention - I told him there was an unusual situation
last month when someone apparently generated some false SSD reader traffic
which to my eyes looked like a possible attempt to emulate the kind of activity
seen prior to an acquisition. I told the company concerned and I didn't write
about it publicly at the time. So if that was the aim - it was thwarted. If
I had been tricked into writing about it I would have passed the details on to
the authorities.
That kind of acquisition fever goes another way too. I
told Ralph I had been talking to a
military SSD company
recently whose name I didn't divulge due to a news embargo at the time -
and I told the marketing conatct at this military SSD company how hard it was
to get any details about their SSD controller architecture - especially as it
played such an important part in almost every new product story which they
were trying to get people like me to write about. I said to that company -
I infer from that your company doesn't have any imminent plans to get acquired,
seek massive VC investments or do an IPO otherwise those controller details
would be all over your website. (This lack of any need for such funding -
because they are profitable anyway - was confirmed by the mil company -
although I would have been happier just to have gotten the technical data I
had been asking for.)
Ralph knew what I was talking about. And most of
you too have seen the kind of activity in private companies where they splatter
their SSD IP all over YouTube as if to say - I've got a great controller - so
please buy my company.
As we had already touched on the topics of
acqusitions and
controllers I asked
Ralph - what did OCZ really think when
LSI said it was
acquiring SandForce?
(October 2011)
Wasn't it a shock having one of your strategic suppliers suddenly turn into a
competitor?
Ralph said that OCZ had already considered the possibility
that SandForce would get acquired - and had even looked into possibly acquiring
it. But OCZ thought the price was too high - for it to have worked out for
them.
That got us onto the subject of OCZ's barefoot controller -
which evolved from OCZ's acquisiton of
Indilinx in
2011.
I
said - I've heard from other sources (readers) that this controller does
incorporate
adaptive R/W
technology - and that's how I listed it - when I published my list of the 10
companies in the market back in the 1st half of 2012. Any other company which
had this type of technology would be saying much more about it - as it's been
such a strategically important technology. But I hadn't seen any original design
architecture details from OCZ itself - and I couldn't find anything about this
parameter on OCZ's web site.
Alex confirmed - yes the current
controller - which is aimed at the
consumer market -
does use adaptive R/W technology. That's what has enabled OCZ to efficiently
customize its SSDs within consumer applications.
Ralph said the main
reason they don't have the controller details publicly on their site is because
its main use is within OCZ's own SSDs. He said it's hard to monetize the
controllers as chip level products. And although OCZ has supplied this
controller to a few strategic SSD companies - OCZ itself remains the most
important user of this design. I also got the impression that OCZ is moving
towards being able to make nearly all its consumer SSDs using its own
controllers.
I wanted to move onto the enterprise SSD market - and
what OCZ was doing there.
I said I've been impressed by the value for
money which OCZ had got when it acquired
SANRAD in
January 2012.
I was trying to figure out what were the characteristics of OCZ's
enterprise customers? - Particulalry those using their PCIe SSDs. I had
guessed - from knowing the market and partly from emails from early customers
- that OCZ was probably getting most of its business from smaller enterprises -
the kind of users which many of the other enterprise SSD companies haven't got
to yet - because it's been easier for them to aim at bigger organizations. As
I've said before on these pages - OCZ is more accessible to do business with -
for many of these small users - because its sales started from the level of
getting a customer even if they're only buying one SSD.
Ralph said -
I was right about this enterprise profile - and he said most of their
enterprise customers are probably in a category which the bigger enterprise SSD
companies would call "tier 2" and "tier 3" users. For these
customers - who have applications at this scale of installation - OCZ's PCIe
SSDs - when bundled with the VXL software (which started at SANRAD) - gives
users almost everything they need in a complete SSD system. Offering what he
called - "A full solution for the smaller enterprise."
I
said - what's remarkable is that OCZ is doing all this business development in
a segment of the enterprise - which when added up is huge - and most of your
competitors haven't touched it yet because they regard it as being too
difficult to reach and not worth their investment yet - so in a few years
time all that marketing experience gained from what you're learning now will
have enormous upside potential. And when your financial levers get freed up
again - your investment in learning this market now will give you great
competitive advantages.
Ralph didn't disagree with that. He cautioned
me that the consequence of having an enterprise customer base with these
characteristics is that each customer might only make a small number of
purchases. On the other hand it meant that if OCZ's enterprise revenue - was
aggregating a lot of these smaller customers - it could be more consistent
and predictable than that of some other SSD competitors who had reported "big
pops of revenue" to a single web company or a big oem. OCZ was starting to
see interest in its enterprise SSDs from bigger systems companies too - but that
was stemming from the fact they had mutual customers.
On the subject
of enterprise SSDs and software
- Ralph said that OCZ has been hiring more firmware engineers to grow their
capabilities in this market.
That got me onto the subject of a new
enterprise controller design - which Ralph and Alex told me about - which will
be a completely new design which also leverages the adaptive R/W technology etc
from their existing barefoot but which is better optimized for high performance
PCIe SSDs. I did learn the date and some of the characteritics - but I won't
say any more about that here - because the details could change and it would be
better for that new controller to star in its own news show.
I didn't
broach the subject of comparing OCZ's new controller design with LSI's new 3rd
generation controller - which was launched this week - because that was under
embargo at that time.
Thinking about it now - I don't think it really
matters whether OCZ's new enterprise SSD controller emerges with chip level
performance specs which are higher, lower or about the same as LSI.
That's
because the key differentiator in any related PCIe SSD would still be the
relative strength and usefulness of the associated software.
For
LSI - the caching
software they offer - is less strategically bound to their controller sales -
because most of their SSD oems will use their own choice of different software.
For OCZ - however - the opposite is true. The things OCZ is
learning from their VXL customer base will make them just as sticky in this
segment of the market as
Fusion-io is in their
own neck of the enterprise - in very large installations. These reasons for
preferring different
suppliers of PCIe SSDs began to be clear a few years ago - and are becoming
more concrete.
At the end of our conversation I thanked Ralph and Alex
for nailing down and confirming a bunch of guesses and also introducing me to
some new insights which point towards OCZ evolving into a much more soundly
based competitor in the SSD market in the next few years than most people
might have imagined.
Although OCZ still has to be super cautious
and prove that it can work as a sound business operating under the financial
constraints it has now - it seems to me that this company - which once operated
in too many markets and was master of none - has been able to judiciously select
some market niches and ways to sustain competitive advantage in them -
especially in the unmined vast regions of the enterprise SSD market -
which could put it a year or so down the track - ahead of the SSD business
learning curve.
...Later:- - 13 days after my
conversation above - OCZ announced it was seeking bankruptcy.
more
details emerge about OCZ and Toshiba
Editor:- December 2, 2013 -
As previously flagged - OCZ
today
announced
that Toshiba
will acquire substantially all OCZ's assets in a chapter 11 bankruptcy
proceeding for $35 million.
Under this agreement Toshiba will acquire
OCZ's client and enterprise solid state drive business. OCZ will continue to
operate and serve existing and future customers during this process.
Toshiba
has agreed to provide OCZ with Debtor-in-Possession financing to ensure that
there is adequate capital and flash supply to support the business during the
contemplated sale period. The consummation of the asset purchase agreement is
subject to an auction and approval by the bankruptcy court in the Company's
bankruptcy case.
This acquisition will provide Toshiba with access to OCZ's proprietary
controllers, firmware and software, as well as the teams responsible for
bringing these solutions to market, in addition to OCZ's established brand and
sales channels.
Editor's comments:- here are my thoughts
which come to mind from this.
- OCZ's VXL software and sales experience in the tier 2, tier 3 enterprise
segment - provide a framework which will enable Toshiba to significantly
escalate its enterprise aspirations - because the skills set and IP from OCZ
will provide Toshiba new channels and tools to market which are independent
of oem design wins.
- If you compare the valuation of OCZ's assets to
prices paid in the
past for SSD software companies and controller companies - this is an order
of magnitude lower than what has been seen in the market before.
This
is probably based on a catious reassessment of just how difficult it is to
plan for revenue growth and profitability - in the SSD market of today - where
there are so many vendors - and in which even heavyweight traditional storage
systems companies like NetApp
have been struggling to get one or two percent share of the enterprise
flash market - and a market in which the tone of revenue projections from
longer established enterprise SSD companies like
Fusion-io and
Violin sound much
flatter than they did a few years ago.
Will this send shock waves
through the VC community?
I doubt it. One of the lessons of the
SSD market in
recent years is that just because something shone from the business point
of view one year doesn't mean the same thing will be good enough the next. New
SSD companies have the ability to disrupt older SSD companies in the same kind
of way that SSDs have been disrupting other linked markets like servers and hard
drives. | | |
|
.
. |
|
|
|
.
. |
|
Here's an easy SSD
question. What does Nimble Storage have in common with Pure Storage? |
Editor:- November 15, 2013 - What does Nimble Storage have in
common with Pure
Storage?
3 guesses allowed.
- Is it - they both have the word "storage" in their names?
Yes but that's not it. And just because we're approaching the weekend doesn't
mean we're going to lower our standards (that far). And if we were (lowering
our standards to that extent) I'd have given you a much bigger list of
companies in my opening question.
- They're both in the advanced stages of doing IPOs?
Yes -
that is indeed so - but please tell me - who isn't nowadays? (Apart from those
who have already done it.) And it's not the answer I was looking for.
- Do they share VCs?
It's a valiant effort... To be honest
I haven't bothered to check. - But even supposing for a minute it were true
- it wouldn't be enough of a unique commonality at this time in the SSD market
- because you know - when it comes to
SSD-VC relationships
we're not talking Jane Austen. VCs are notoriously promiscuous. (And
whatever happened to only 3 guesses?) OK - you give up - here's the
answer I had in mind...
Both Nimble and Pure are being sued by old
style computer companies who assert that when their employees joined these
irritating startups they took with them customer-centric market knowledge.
I already discussed the Pure case before so I won't repeat that here.
But although you may have known about Nimble too - known indeed days
or weeks ago - in which case congratulations on having won the quiz! - sorry
there are no prizes - and no I'm not at all offended that you read other SSD
sites too - especially if they're on the subject of
consumer SSDs -
not my shade of purple don't you know? - I have to confess that I only learned
about Nimble's pursuit by NetApp as late as this morning - and my sorry
excuse for having missed this story is I've been busy on other SSD stuff
- which you'll hear about later.
On the other hand if like me - you
didn't win the quiz despite coming to this site every day and never reading
anything else - thanks - mea culpa - normal service will be resumed as soon
as possible - and here's one of many
links for
the the Nimble/NetApp story.
Now... another thing Pure and Nimble
have in common is names which suggest implicit virtues.
I've done a
bunch of articles already on many of the aspects of
naming SSD brands.
Speaking
of which - if some product marketer out there decides to call their new product
line the "good enough - but no so good you can't afford it" enterprise
SSD range - let me know and I'll update my SSD name lists.
Industry
experts and analysts might say the "good enough - but no so good you can't
afford it" enterprise SSD line already exists - but goes by
aliases.
Why's that? Because the marketing positioning gurus or the
company's founder or the vp of sales didn't think that "good enough"
was good enough. And BTW the CFO said - have you any idea what this would cost
on Google adwords?
But I
have reasons to suspect that a skunks work SSD marketer who works in
SEO hasn't dropped
the "good enough" SSD brand idea entirely and is spending a lot of
time looking at Thesaurus.com and
synonymically related sites - while also holding another window open on
SSD jargon. | | |
|
. |
.. |
|
|
|
|
. |
|
|
|
.. |
|
|
|
.. |
|
|
|
.. |
|
Surviving SSD
sudden power loss |
Why should you care
what happens in an SSD when the power goes down?
This important design
feature - which barely rates a mention in most SSD datasheets and press releases
- has a strong impact on
SSD data integrity
and operational
reliability.
This article will help you understand why some
SSDs which (work perfectly well in one type of application) might fail in
others... even when the changes in the operational environment appear to be
negligible. |
| | |
|
. |
|
| |