|
|
|
| .... |
|
|
Skyera is one of the world's top 4 most important SSD
companies (ranked by search metrics of 200,000 approx StorageSearch.com
readers in the 2nd quarter of 2013.)
The company's primary products
are
rackmount SSDs
which leverage Skyera's IP leadership in
SSD design efficiency,
adaptive R/W
flash, big
controller architecture and
SSD software.
A
reader recently asked me about the significance of Skyera's high ranking in our
SSD reader searches in the light of what he referred to as "Skyer having
lots of ideas… but not many sales."
Generally I referred him
to the proven past successes of the 6 year running
Top SSD Companies Lists
as a sensitive and early indicator of future business success or failure - an
issue which I discussed with noteworthy historic examples in the
24th quarterly edition
(published April 2013).
Specifically - re his question - as to why so
many of our readers might be interested in learning more about Skyera I said -"Nearly
everyone in the enterprise systems business I talk to (CEOs , CTOs etc) agrees
that Skyera will be the company that sets the agenda for the price of
enterprise SSD racks for the next 3-4 years. Everyone else either has to
figure out a way to beat them in niche segments or avoid competing with them
altogether."
see also:-
exploring the
exciting new directions in rackmount SSDs |
| .. |
|
|
In August 2011 - in
a
white
paper presented at the Flash
Memory Summit -
Rado Danilak,
founder of StorCloud
said that RAID / RAIS
technology and the other clever
SSD controller
techniques that the industry has learned to appreciate in the past few years
"...will not sufficient for advanced node MLC to be used in
enterprise... But - StorCloud has identified technical solutions for next 10x
improvement in
endurance
and enabling 1x nm MLC
in enterprise storage systems..."
In April 2012 -
StorCloud changed its
name to Skyera.
In
June 2012
- Skyera moved into
its new (81,000 square feet) headquarters at 1704 Automation Parkway, San Jose,
CA 95131.
In August 2012 -
Skyera
launched its first product - a 1U half-depth 10GbE
SSD rack with 44TB usable capacity ($131,000 approx), 3.6GB/s throughput and
upto 1 million IOPS using under 800W electrical power. Integrated software
includes inline dedupe, compression, snapshots, clones and various management
functions.
In February 2013 -
Skyera
announced
$51 million in financing led by
Dell Ventures and including an
investment by WD.
In
August 2013
- Skyera
announced
that it will introduce
PCIe connectivity (to
the existing FC and
iSCSI) as well as
replication in the next version of its
rackmount SSDs -
the skyEagle - which will ship in the first half of 2014 - offering 500TB
uncompressed (2.5TB deduped and compressed) in a 1U form factor at a record
breaking list price
expected to be under $2,000 per uncompressed terabyte. |
| .. |
|
|
| .. |
| Skyera
launches industry changing enterprise SSD |
Editor:- August 14, 2012 - Skyera today
launched its first product - a 1U half-depth 10GbE
SSD rack with 44TB usable capacity ($131,000 approx), 3.6GB/s throughput and
upto 1 million IOPS using under 800W electrical power.
Integrated
software includes inline dedupe, compression, snapshots, clones and various
management functions.
Skyera will demostrated the new
Skyhawk next week at the
Flash Memory Summit.
Editor's
comments:- last week I spoke to Skyera's founder and CEO Rado Danilak
and some key people in his team - which in addition to core members of the team
who designed some of the early
SandForce controllers -
also includes talented and experienced
people from other well
known chip and enterprise storage companies.
Skyera make a big thing
about their cost value proposition.
They deliver fast SSD storage
at a system list price under $3 / GB (uncompressed) and about $1 / GB (with
compression and dedupe). That not only beats best of breed SSD systems but
seriously challenges HDD based storage too.
Designed from the ground
up the Skyhawk can be summarized as using
adaptive DSP,
with big
controller architecture (proprietary rather
than an array of COTS open SSDs) and with an integrated
software stack which
screws the efficiency of
reliability and
data integrity
as tight as can be - by intervening at multiple levels (high
adaptive
intelligence flow symmetry.)
In any given year in the SSD market
there are maybe 2 or 3 industry changing announcements - which illustrate what
the shape of the future could be.
Skyera's Skyhawk is one of those
in the year we're already in. And would be just as significant if the launch
had been delayed into next year too.
Here's some more stuff I learned
from talking to Skyera.
Funding
Rado said Skyera has
one strategic investor - which he didn't want to name right now. The company is
mostly owned by its management. That gives them an incentive to design a product
which can make money as soon as possible.
FPGA rather than ASIC -
for Skyera's hardware controller
The FPGA versus ASIC tradeoff
debate in electronic design goes back to the mid 1980s. With FPGA you buy a
chip that's got arrays of logic and common functions already in it. You
define what your FPGA does by programming the interconnections in a way
that's similar to writing to flash. With ASIC you have almost total freedom to
decide what functions your chip does - but the chips are made in a factory
that's making a lot of other chips - not just yours. You're buying time in
someone else's multi-billion dollar factory. There are set up fees, learning
curves and queues involved.
Skyera implements its controller using
FPGAs.
That gives flexbility (also
explained to me as
a factor by Holly Frost at competitor
Texas Memory Systems
- which uses both ASICs and FPGAs in different places - while
Nimbus prefers to use
exclusively its own ASICs and
Fusion-io is still
staying mostly with FPGA.
In a rack system - the recurring unit cost
is not a big deal - whereas it is a
cost disadvantage
in a single small drive - which is why
STEC made much of
switching to ASICs in its enterprise SSDs last year. Where the VC factor comes
back into the equation is that each design costs hundreds of thousands of
dollars - and months of delay. So if you have the realistic choice of either
FPGA or ASIC in the form factor and power budget - which is not an option if
you're designing a 1 inch
embedded SSD for example - then the FPGA route means you can ship products
sooner with less initial outlay and your revenue stream can contribute to
earning profits faster.
Having said that - my guess is that after
their launch - if Skyera said they needed another $500 million or so to do
something - like remodel the executive car park - VCs would rush in to supply
the cars too. But Skyera probably doesn't need that much money.
Core
SSD technologies
On the SSD architecture side I've already written
about the strategic importance of large architecture, adaptive DSP, SSD
software, tradeoffs
between flash capacity and reliability and performance and the benefits of
integrating them at a high level. So I won't repeat those points here.
Rado
said he wanted to get the best performance and reliability possible out of "cheap
crappy flash". He said if you use a traditional
RAID 6 array - then the
latencies start to mount up - and you end up doing 3 writes cycles to get the
parity protection. That's bad for endurance and bad for latency too. He said
that Skyera's way of doing RAID-like protection means that the write
amplification for a 4kB write is an average of only a factor of x1.04.
That's a 3x reduction on write amplification compared to arrays of
2.5" SSDs for
example. There are many different levels in the hardware and software - where
the self knowledge about the system is used to good advantage.
Their
controller technology can be adapted to work with any flash. In theory they
could populate production systems with the cheapest flash they can get and
even mix up the sources in the same SSD and still get a working system.
Nevertheless the company's deep knowledge of flash - means that (in my words)
there's little risk of cooking the memory cells to get the performance. And they
know which suppliers are better.
Earlier this year
DensBits explained
to me how using adaptive DSP meant that you could get 60% faster write
performance out of MLC flash because the DSP gave you the data integrity when
using shorter write pulses and lower energy - which also leads to better
endurance.
(See endurance
multiplication wars.)
Anyway Rado explained - that once they
decided that they had to design a proprietary memory array - they knew they
also had to invest in the software layers to make it work within the enterprise
infrastructure.
Internally their box is like an array of memory
connected to a switch. In that respect - it's similar to the RamSan systems from
TMS.
High
availability
Skyera's initial system isn't a
true high
availability SSD - and although it has many essential enterprise features -
this will limit the appeal of the product in some applications. Nevertheless, I
know that many super users of SSDs would prefer to use their own cloud like
wraparound software - and all they need is a low cost scalable rack which they
can use as a component to build multi-petabyte SSDs.
Skyera said they
see traditional enterprise SSD apps like analytics being a good fit for the
launch system.
HA is in the product roadmap and has hooks in the system
architecture. It will come later.
Business development
I
suggested to Skyera - that given how
popular their
company had been with SSD readers while they were still in stealth mode - the
problem they might have after this product launch would be too many inquiries.
That would place a burden on marketing to filter out the type of SSD leads
which were a good fit and those which were just curiosity.
Maybe we can
recycle the leads we don't want and sell them to other SSD companies - said
Skyera.
The SSD market will be watching Skyera with interest to see
if stacking all these new technologies together for the first time in a
system really works. | | |
| .. |
 |
| . |
StorageSearch.com's
original profile of Skyera
the emerging from stealth edition -
August 14, 2012 |
Skyera (which was initially called
StorCloud) today emerged from stealth mode.
The company's founder -
Rado Danilak
was previously the founder and CTO of
SandForce.
Skyera's product is a rackmount
SSD. It doesn't yet have true
high availability
- but its internal data integrity and chip optimized RAID - are good enough for
a wide class of enterprise uses today.
Like
Violin and
Texas Memory Systems -
Skyera has a big
SSD controller architecture.
But unlike those 2 companies - Skyera
is using (in Rado's words) "cheap crappy flash" - which it brings into
line with many integrated design techniques - including
adpative DSP
write management (similar to
SMART,
STEC and
DensBits.
Like
Nimbus Skyera has its
own software stack.
And like
Pure Storage -
Skyera integrates inline compression and dedupe as features which can push the
cost per SSD terabyte even lower (below $1/GB at time of product launch.)
Skyera's
solution will be attractive to users looking for scalable and cost effective
multi-petabyte SSD
systems.
But its
proprietary - rather
than open (array oif COTS SSDs) approach will limit the company's initial
uptake with more conservatively minded users - who would prefer to see an array
of SAS or
PCIe SSD modules
inside a rack.
However, the low latency which is achieved by using a
proprietary array design and the efficiencies (in cost and reliability) which
accrue from integrating many leading edge SSD design techniques concurrently in
a reinforcing manner within a single rack system - are just not possible using
so called "open" designs.
As with any new generation of
SSD controller - which
pushes the boundaries of what you can do with cheap memory - there are risks -
because the product reliability and life cycle and
endurance
model is based on extrapolated predictions and models - and isn't yet market
proven.
That should bring some comfort to Skyera's competitors in
the rackmount SSD market in the next few quarters. | | |
| . |
|
|
| . |
|
|
| .. |
 |
| |
.. |
"You may decide that
my ratios are too timid"
I said to Skyera's CEO - Rado Danilak
"if
so - scare us!" |
| scary Skyera - I
talked to Skyera's CEO about SSD utilization efficiencies and the problem of
being a newbie vendor in the enterprise SSD market. (October 22, 2013) | | |
| .. |
 |
| .. |
| Skyera says 1 petabyte SSD
will go into 2U |
Editor:- August 13, 2013 - Skyera today
announced
that. among other things, it will introduce
PCIe connectivity (to
the existing FC and
iSCSI) as well as
replication in the next version of its
rackmount SSDs -
the skyEagle - which will ship in the first half of 2014 - offering 500TB
uncompressed (2.5TB deduped and compressed) in a 1U form factor at a record
breaking list price
expected to be under $2,000 per uncompressed terabyte.
Editor's
comments:- not a lot of people remember this - but 6 years ago- in
August 2007 -
when Violin emerged from
stealth mode with one of the
fastest rackmount
SSDs of that era - Violin's 1010
was also the first well known enterprise SSD which offered PCIe as the
primary connection option. (This was the month before
Fusion-io launched
its first ioDrive and began its multi-year mission to re-educate and change the
way that the enterprise market viewed SSDs BTW.)
But in
2007 the
enterprise market was still grappling with the idea of rackmount SSDs - and for
those buying - their connection of choice was
FC SAN. That meant
Violin's initial product made a market impression - but the company had to
wait for its later systems - redesigned with FC and flash - to get the sales
which would secure its future as the new leader in the rackmount SSD market
(replacing Texas Memory
Systems). Violin's early experience with PCIe being the wrong interface for
its rackmount market left scars in the company's psyche. So it wasn't till
nearly 6 years after that Violin entered the (now) safely conventional market
for module and card based PCIe SSDs.
Today in the market of August
2013 - PCIe has evolved into a very different technology and market
proposition for enterprise SSDs.
- As a technology (it can be used as a fabric to interconnect fast racks - a
bit like InfiniBand
in some respects - except that the ecosystem of compatible SSDs is much richer
and price competitive).
So
I view the 96 lanes of PCIe connectivity - in Skyera's forthcoming skyEagle -
as a way of reaching out to an entirely new market. This PCIe option provides a
simple and cheap foundation for clustering boxes in
high availability
configurations.
And how about the idea of 1/2 petabyte SSD in a 1U
rack next year?
Back in early
2010 - when
I published an article designed to show what one of the missing products in the
future enterprise SSD
market would look like - and its consequences for
hard drives -
this way to the
petabyte SSD - I guessed we'd see a 2U petabyte SSD in 2016.
Skyera is going to beat that timescale by 2 years. Which incidentally means
that all the other predicted dates about the consequences for
hard drives in an
enterprise SSD world - are also 1 or 2 years earlier than anyone previously
imagined too. | | |
| .. |
 |
| .. |
| Skyera unifies 19nm MLC
arrays with 100x life |
Editor:- May 21, 2013 - Skyera today
announced
it has added unified storage operation (concurrent NAS and SAN) to its
pre-existing SSD box
(Skyhawk).
Editor's
comments:- this was already anticipated and factored in by potential
systems competitors that I've spoken to in the past several quarters.
More
interesting for me - is the "100x MLC life amplification" figure
quoted in
a
recent blog by Skyera's CEO.
When you're asking what's possible
from combining controller
technologies (like
adaptive R/W)
with software efficiencies
(don't do things which are unnecessary to access the true app data - as opposed
to emulating every just-in-case-we-need-it lookahead or spurious
hard drive traffic
request) the 100x figure is a useful competitive metric and shows what
you can integrating array level system software with already optimized drives.
It's all about being at the leading edge of the system
SSD price curve.
See also:-
MLC Seniors live longer
in my SSD care home | | |
.. |
| WD invests in Skyera |
Editor:- March 12, 2013 - WD was one of the recent
investors in Skyera
- it was
announced
today.
We see companies like Skyera as offering a dramatic improvement
over traditional approaches to emerging storage challenges said Steve
Milligan, president and CEO, Western Digital. | | |
| .. |
 |
| .. |
|
|
| .. |
| "Although few
people on this planet really understand the complex mix of technologies which
Skyera has mastered to architect one of the world's most efficiently
engineered SSD arrays - almost anyone can easily appreciate the results when
they're presented with the resulting price and performance." |
| Editor:- January 11, 2013 -
the Top SSD Companies
in Q4 2012 | | |
| .. |
| "...the most
significant emerging technology in 2012 was adaptive R/W DSP flash
management...." |
| Strategic Transitions
in SSD - (December 20, 2012) | | |
| .... |
| Skyera says
a little more about its SSD software |
Editor:- October 16, 2012 - Skyera today
revealed
a few more features of the software which supports its
recently launched
rackmount SSDs - along with an overview
and some pretty pictures.
Among other things - which were new to
me - administrators can allocate LUNs according to 3 different classes of SLA
for capacity and performance.
"A true solid-stage storage solution
must be more than simply sticking flash media and controllers in a box"
said Skyera's founder and CEO Rado Danilak.
Editor's
comments:- as I said in the
enterprise
SSD survivor's guide - "Software used to be SSD's enemy. Now it can be
SSD's best friend."
See also:-
SSD software,
SSD design efficiency
| | |
| .. |
|
|
| .. |
 |
| . |
|
|
.. |
|
|
. |
| other SSD
articles worth seeing |
The
top 100 SSD
articles and blogs (updated monthly) is based on reader popularity.
The
beginnning of the SSD search is a different place for everyone. Just as every
day someone is hearing the music of the
Beatles for the very first time
- so too do some old classic SSD articles and themes seem to endure.
But
if I had to suggest just 3 SSD articles - depending on who you are - it would
be these.
for experienced enterprise SSD readers
- where are we
now with SSD software? - and how did we ever get into this mess? In the next
few years the software for SSDs will have as much impact as hardware
architecture did earlier. Don't expect this to be an orderly top-down process.
if you're new to the SSD market
- the SSD Buyers Guide - will take you to groups of
articles and directories which are organized by technology and market theme.
- the Top 20 SSD
Companies - whatever your interest in SSDs - these are the companies which
are attracting the most interest amongst your peers.
| | |
| . |
| |
