 |
|
|
| leading the way to the
new storage frontier | |
| ..... |
Companies
mentioned this month include:-
Diablo,
Crocus,
EMC,
HGST,
Micron,
OCZ,
PLX,
Samsung,
SanDisk,
ScaleIO,
SMART,
SolidFire,
Stec,
Tintri,
VeloBit,
ViON,
Virident,
Virtium,
WD,
WhipTail
See
also:- SSD
market history,
the SSD
bookmarks,
latency
reasons for fading out DRAM |
|
 |
| .... |
industrial grade mSATA SSDs
from
Cactus Technologies | |
| . |
Diablo discusses details of
Memory Channel Storage
Editor:- July 30, 2013 - Diablo Technologies
today did the
public launch of
its its new technology -
Memory Channel Storage
- which repurposes the interface and form factor of server DRAM into a new
architecture for ultrafast flash SSDs which the company positions as a
competitive alternative to very fast
PCIe SSDs.
Editor's
comments:- I spoke recently to Diablo's new VP of Marketing - Kevin Wagner about the
company's new MCS.
I had already gleaned a heads up on the boundary
capabilities and market potential of the new technology in
an earlier
interview with Diablo's flash partner SMART. So it was natural to ask how
Diablo viewed the impending acquisition of
SMART by
SanDisk?
Not
surprisingly Kevin indicated that whereas Diablo already thought it had made the
best choice before - the upside potential of having SanDisk as a partner for
this memory technology made it 10x better.
Here's what I
learned.
- Diablo's TeraDIMMs are designed to fit standard sockets designed for DDR-3.
They are electrically, form factor and power compatible. But instead of RAM - a
typical TeraDIMM using today's technology might have 400GB of flash.
- Diablo's controller architecture means that the host CPU can read and
write from memory in the same transparent way as it would talk to RAM.
- TeraDIMMs can be installed in every set of slots where you'd normally
insert RAM. The only limiting factor in the current architecture is that at
least one pair of slots has to be RAM. All the rest can be flash.
- Diablo claims that MCS has better write latency than most PCIe SSDs.
Specifically MCS has a write latency of 3 to 5 microseconds. This is
really a write to the controller.
TeraDIMMs have a
skinny flash
cache
architecture - which means that apart from a small amount of register
memory in the MCS controller itself - no
RAM is used in the SSD.
The MCS design includes
power fail
hold up capacitors which guarantee that all data which has been written to
the TeraDIMM gets completely saved to flash.
- From the applications point of view MCS looks like a massive amount of
persistent RAM - but with terabytes rather than tens of gigabytes of memory
space and with a cost structure closer to the
market price of flash
than DRAM.
- Is it bootable? No. Not yet that wasn't regarded as a priority.
Although
the product is not yet available - Kevin Wagner told me that Diablo has been
getting a lot of interest from server companies. Diablo has been validating
software with their ASIC based implementation for several months following good
earlier characterizations of the design with FPGA. From the software point of
view Diablo's aim was to prioritize a usable design which would work for the
market as soon as possible. Obviously many possibilities for leveraging the
basic technology spring to mind. I asked for example about preferred models of
high availability?
Kevin said that so many companies are interested in
what they're doing that they don't have enough resources to talk to them all
right now. It won't be long before the company publishes more details of its
reference architecture - and conversations have already begun with ISVs and
other companies which could be the seeds of a new ecosystem.
But the
current design has been designed to work in virtualized and non-virtualized
environments and the company has got already got a good idea of how headline
big data apps would perform in benchmarks using their technology.
I
raised the question of inter-operability with
PCIe SSDs (in the same
MCS resident servers) and it looks like the guesses I had made about
segmentation and collaboration and competition with other SSD types which I've
already written
about in my earlier article about memory channel SSDs - are still valid. So
I won't repeat those points here.
We still have to wait for firm
product pricing and configuration details. But if you had any doubts about where
the money will be inside servers - MCS provides another new way of packing
even more flash SSD capacity in.
Samsung invests in SolidFire
Editor:- July 25, 2013 -
SolidFire
today
announced
it has raised $31 million in series C funding which includes a new investor
- Samsung Ventures. SolidFire's
latest product - the SF9010
- an iSCSI compatible
fast-enough
rackmount SSD -
which will ship in September - uses SSDs from
Samsung.
See
also:- VCs & SSDs,
exciting new
directions in rackmount SSDs,
hostage to the
fortunes of SSD
WhipTail announces new channel for defense customers
Editor:-
July 24, 2013 - ViON
announced it will now serve as an authorized provider of maintenance and
support services for the entire WhipTail product
line.
"This partnership took off primarily due to the great
success at WhipTail with the
defense and
intelligence communities and ViON's clearance and track record of successfully
providing first level support for other vendors." said Dan Crain,
CEO of WhipTail.
Tintri announces record breaking business milestones
Editor:-
July 23, 2013 - Among other things - Tintri today
announced
that in the 1st half of 2013 its bookings were over 2x the level they
had been the year before.
OCZ ships PCIe SSD based SQL accelerator
Editor:-
July 23, 2013 - OCZ
today
announced
the general availability of its
ZD-XL SQL
Accelerator - an SSD
ASAP appliance - delivered as a PCIe SSD (600GB, 800GB or 1.6TB) and
bundled software - which optimizes caching of SQL Server data in Windows
environments - and can provide upto 25x faster database performance.
HA
functionality works through Microsoft SQL Server AlwaysOn technology, so that
in the event of planned or unplanned downtime, can continue operations from the
stopping point, retaining all of its data as if no downtime had occurred.
"We believe that the industry is primed for this type of tightly
integrated, plug-and-play use-case acceleration solution..." said Ralph Schmitt,
CEO - OCZ Technology.
Editor's comments:- One of the
differentiators in SSD caching products is the sophistication of their
behavior when viewed from a time basis. This is 1 of the
11 key SSD
symmetries - which I call "age symmetry".
In this respect
- a key feature of ZD-XL SQL Accelerator is its business-rule pre-warming
cache engine and cache warm-up analyzer that monitors SQL Server workloads and
automatically pre-loads the cache in advance of critical, demanding or important
SQL Server jobs. It achieves this by identifying repeated access patterns that
enable DBAs to set periodic time schedules to pre-load the cache.
This
product won Best of Show Award at an event called Interop in
May.
Industrial Grade eMMC - new blog by Datalight
Editor:-
July 23, 2013 - Datalight
has published a new blog -
What is
Industrial Grade eMMC? - written by Roy Sherrill,
President - who says - "Industrialized eMMC is gaining consideration for
embedded applications in areas such as automotive, medical, aerospace, and other
commercial applications."
See also:- more blogs by
Datalight - re extending embedded flash life using software and
tiny SSDs and SSDs on a
chip
Samsung enters the 2.5" PCIe SSD market
Editor:-
July 18, 2013 - Samsung
today announced its entry into the
2.5" PCIe SSD
market. Its new NVMe SSD has upto 1.6TB capacity, read throughput upto
3GB/s, and up to 740K
IOPS
Overview of PCIe topologies for enterprise SSDs
Editor:-
July 17, 2013 - PLX
Technology recently published a white paper -
Enterprise Storage
and PCI Express - which gives an overview of past, current and future PCIe
SSD connection topologies along with a list of detailed reference articles.
Along
with the clear text and authoritative viewpoint (PLX is the leading supplier of
PCIe chips to the SSD industry) the article includes a complete set of
attractive pictures too - which make it easier to visualize the many different
interconnection methods being discussed. Below is an example from a section in
the paper - called - Taking PCIe Out of the Box.
In recent years we've seen the start of a growing
diversity in both
the type and functionality of
PCIe SSDs. PLX's new
article provides a good introduction to what can be done with PCIe in an SSD
context - and may make you rethink your ideas about the roles of this interface
too. ...read
the article
See also:-
interface chips and IP,
7 silos for enterprise
SSDs
EMC's acquisition of ScaleIO points to SSD server future
Editor:-
July 16, 2013 - EMC
recently
announced
it has agreed to acquire another storage software company - called ScaleIO.
EMC indicated that
ScaleIO's software - which emulates the capabilities of virtual SAN style
storage within the physical implementation of pools of server attached DAS
- makes it easier for users to manage expanding data volumes and reduces the
need for performance planning. The new software will be applied to extend the
application functionality of EMC's
PCIe SSD product lines
and XtremIO rack based flash systems.
Editor's comments:- One
way to view this is it will give EMC similar capabilities to
Nutanix. Or another is
that the EMC/ScaleIO solution (if and when it's done) can be seen as a shot
back across the bows aimed at
Fusion-io's ION
software. (You came into our market space - so we're coming into yours.)
Take
a step back however, and it doesn't have to be so personal.
Most legacy
systems have shapes and architectures which date back to a command and control
SAN style architecture
dating back to the 1990s.
If you were trying to solve the same data
processing and content management functions from a clean sheet start today -
you'd probably go for a more "democratic" Google style architecture -
in which most racks in the datacenter are similar - and their function is
defined and can be changed by software - rather than being hardwired by the
description of the box at the time it was invoiced.
It's long been
known that SSD acceleration lets you speed up legacy architectures - but SSD
performance also gives you the freedom to emulate entire applications
environments on cheaper, and more
efficient, modern
hardware.
Micron
samples 16nm nand
Editor:- July 16, 2013 -
Micron today
announced
it will be in full production of 16nm nand flash (128Gb MLC memory devices)
in Q4 this year - and is designing SSDs around this process geometry - to ship
in 2014.
See also:-
an SSD view
of semiconductor memory boom bust cycles
OCZ is now an enterprise SSD company
Editor:- July
15, 2013 - OCZ's
quarterly enterprise SSD revenue has grown to over $25 million - approximately
half of its total SSD revenue - the company
announced
today.
Editor's comments:- from a revenue perspective
OCZ's situation is way down compared to the situation it had predicted in
May 2012 - when
the company reported that its revenue for the year ending February 29, 2012 was
$338 million - and at that time OCZ was suggesting that its revenue for the
following year would be over $600 million.
That didn't happen for
reasons reported earlier - partly due to errors in accounting but also due to
better judgement - and decisions to withdraw and downsize sales in the
most unprofitable segments of the SSD market - which at that time were
consumer SSDs.
OCZ's
latest results show that it is possible for a company which started out
as a consumer SSD company to change itself into being a (mostly)
enterprise SSD business. (Although it wasn't quick, painless or easy.)
That's
an identity switch which some other significant SSD companies would also like
to do for themselves too.
the Top SSD Companies in Q2 2013
Editor:- July 11,
2013 - StorageSearch.com
yesterday published the
25th qiuarterly
edition of
the Top SSD Companies
List - based on search metrics in Q2 2013.
One of the big changes
I've seen recently is that SSD oems have learned that selling SSD components in
the enterprise market is a risky business strategy - because they don't learn
so much about user needs at the array level, find it harder to perform system
wide optimizations and have less direct contact with end-users. Also some of the
most competive products in the enterprise market aren't simply built from arrays
of commercial off the shelf SSDs.
When the
Top SSD Companies series
started over 6 years ago - 8 out of the top 10 SSD companies marketed SSDs in
traditional hard drive
form factors. In the intervening years we've seen the emergence of many other
important trends - such as the rise of
controllers and new
markets created by PCIe
SSDs.
Now there's another new trend identified in the
latest edition. 8 out of the top 10 companies market rackmount SSDs. ...read the article.
Virtium offers 64GB SLC TuffDrive CF for "Godfather"
slots
Editor:- July 11, 2013 - According to Gary Drossel,
VP of product strategy at Virtium - "There
has been a misperception in the industry that, because other suppliers have
exited the market either through EOL or industry consolidation, CompactFlash
(which has been
the
Godfather of embedded form factors (pdf) for more than 10 years - ed) is
no longer viable as an SSD technology in the
industrial SSD
market. That is absolutely not the case."
Supporting the
view that as long as there's enough demand for legacy form factors someone
will supply it - Virtium recently
announced
a new generation of high density industrial SSDs (upto 64GB SLC) in its
TuffDrive
CF range which meet MIL-810 standards for shock and vibration.
HGST catches VeloBit
Editor:- July 10, 2013 - For the
past 15 years from what I've seen - the ultimate business aim of most storage
software companies has been - to get
acquired.
That's
been even more true in the SSD
software market - wherein frankly - most companies don't even pretend to
invest in sustainable business models.
In the past 2 years - an SSD
software company has been
acquired every 2
months (on average) and the latest company sustaining that trend is VeloBit which has been
acquired by WD
for deployment by its subsidiary HGST - it was announced
today.
In case you've forgotten why this trend started - software
makes it easier to sell more SSDs and the ROI from a vendor's point of view is
better than doubling the sales force. That's why valuations (not disclosed in
this case yet) have been so disconnected from the financial outlook of the ISV's
themselves. See also:-
SSD ASAPs
Crocus gets additional $45 million funding
Editor:-
July 10, 2013 - Crocus
Technology today announced it has raised $45 million (approx) in
additional capital. A new private investor, Industrial Investors, joined the
round led by Idinvest Partners. See also:-
industrial SSDs,
nv memories,
SSD controllers
Virident names SAS SSD veteran as new VP Engineering
Editor:-
July 9, 2013 - Virident
Systems today
announced
the appointment of a new VP of Engineering - Mark Delsman - who had
formerly held similar positions in both SanDisk and Pliant. See
also:- SAS SSDs
Stec makes changes in defense management
Editor:-
July 9, 2013 - Stec
yesterday
announced
the appointment of a new VP for Government and Defense related aspects of its
business.
Editor's comments:- This is not entirely unexpected.
2 weeks ago in my comments about the WD-Stec acquisition news I said that
spinning off the legacy defense SSD business might be a good long term option as
I couldn't see a good fit with that and HGST's other activities. Whatever does
transpire - it's got a separate value and values. See also:-
military SSDs
Hostage to the Fortunes of SSD
Editor:- July 3, 2013
-
Hostage to the Fortunes of SSD - is
the title of the new home page blog here on StorageSearch.com
In a
lot of the conversations I've been having recently with SSD companies and
investors I've found myself trying to explain what are the underlying
pressures which compel rational companies to pursue what appear to be
irrational SSD business strategies. They aren't free to choose alternative
realities. ...read the article
why is SanDisk acquiring SMART Storage Systems?
Editor:-
July 2, 2013 - SanDisk
- today
announced
a definitive agreement to acquire SMART Storage Systems
for approximately $307 million. SMART's revenue in the quarter ended May 31,
2013 - was $25 million.
The transaction is expected to close in August
at which time approximately 250 employees of SMART will join SanDisk.
"SanDisk is excited to build upon its leadership position with
its fourth acquisition in the enterprise storage market," said Sumit Sadana, executive
VP & chief strategy officer of SanDisk. "This acquisition enables
SanDisk to address a $1.6 billion market opportunity in enterprise SATA
products, and complements our strong enterprise SAS product portfolio. With this
combination, SanDisk will have products qualified with 6 of the top 7 storage
OEMs worldwide."
Editor's comments:- Earlier this year year I said to some
readers that if someone wanted to acquire world leading adaptive R/W flash
controller technology which is suitable for the enterprise - and a
SAS SSD product line
which has been adopted my many leading oems - they would get a better deal
acquiring SMART rather than Stec.
(And less mess to untangle too.)
What's this acquisition about?
To
understand the key elements involved - see these articles.
- the competitive advantages - See my article about
efficiency in SSD
architecture.
Apart from short term competitive advantages - the
strategic importance of efficiency for a flash maker like SanDisk - in a 2-3
year window where no-one wants to build new wafer fabs until they're sure about
the next steps in nv memory - is that given the same number of flash memory
chips they can build maybe 30% or more enterprise SSDs - and also use lower cost
memory - compared to standard model controllers.
- the price. - Recent
acquisitions in the
SSD market and the general tough competitive outlook for any single SSD
company seem to indicate that the realizable value of any privately owned
SSD companies has been dampened by 50% or more compared to the peaks of
earlier years.
In this environment it's better for
VCs - such as SMART's
investor Silver Lake to sell now and
cash out rather than wait years for virtual valuations to get back to a peak
they have already past. | |
| . |
 |
| . |
|
| |
|
|
| .. |
 |
|
SSD news
SSD market history
Can you
trust SSD market data?
How fast can your SSD
run backwards?
soon all enterprise
data will touch an SSD
understanding flash SSD
performance limitations
|
|
. |
|
how new SSD software
gets things done faster
by Zsolt Kerekes,
editor - July 29, 2013 |
One of the latency reducing tricks in a
world where every SSD vendor has access to the same
flash memory and
interface chips and choice
of controller
architectures
is the applications magnifying power of
SSD software.
Sometimes
the way that new SSD software gets things done faster is to avoid doing
some things at all - by carefully discriminating between - what needs to be
done - compared to what would normally get done in blind obedience to
tradition.
One of the ironies of legacy systems software running in
flash systems is the way that the data weaves through layers of fossilized
unreality where emulation is stacked on emulation - and hardwired into
the software and data flow logic are the remembered
once-deemed-to-be-efficient solutions to data flow control problems whose
origins are now almost forgotten.
So the SSD emulates a hard drive.
And the hard drive emulates memory.
And it gets worse.
The fetching and prefetching and polite but useless flurries of activity which
happen behind the scenes makes it appear more like a bunch of courtiers in a
fairy tale palace reacting to this simple request.
The Princess needs
shoes.
What shoes? What color? What style? What for?
She
hasn't said yet - just get as many shoes as you can carry and be quick about it!
Yet despite all this background mayhem the application - somehow -
still runs faster on SSDs than on the old hardware. (And the Princess has never
been seen in public without wearing appropriate footwear.)
The other
way to save time (improve latency) is to say - what if instead of just
speeding up all the tangled processes of emulating a hard drive emulating
memory and worrying about all the old fossilized limits of packet sizes and
flow control in drives and interface cards which no longer exist except in
museums but which have been preserved in legacy software - we instead
make an effort to write some new software which knows it's operating in a flash
world and doesn't have to recite old HDD spells to charm the data?
Or
what-if the Princess knows where the shoe room is - and rather than wait -
she's going to get the shoes for herself?
The implications of these
what-if? results (for SSD software) are easy to anticipate and we've seen what
happens when these ideas have
found their way
into SSD benchmarks but
it still takes time for these new ideas to work their way into standard software
products.
And if the Princess changes her mind between the time she
sets off to the shoe room and when she gets there - she's still going to get the
shoes she wants quicker than
if she asked her maid.
All of which is a preamble to say that Fusion-io last week
announced
that its Atomic Writes API contributed for standardization to the T10 SCSI
Storage Interfaces Technical Committee is now in use in mainstream MySQL
databases MariaDB 5.5.31 and Percona Server 5.5.31.
Modern SSD
Princesses prefer not to be kept waiting. | | |
|
. |
|
 |
|
. |
|
SSD endurance - the
forever war
how
long for hard drives in an SSD world?
exciting new
directions in rackmount SSDs
Data Integrity
Challenges in flash SSD Design
Adaptive R/W and
DSP ECC in flash SSD IP
Efficiency - making the
same SSD - with less chips
MLC flash lives 10x longer in
my SSD care program
how will Memory
Channel SSDs impact PCIe SSDs?
|
|
. |
|
 |
|
. |
|
Where are we now
with SSD software?
How will hard drives
fare in an SSD world?
exciting new
directions in rackmount SSDs
SSD ASAP / auto tiering SSD
cache appliance news
comparing SSDs to
other disruptive changes in computer history
|
|
. |
|
| the Top SSD Companies in
Q2 2013 |
Editor:- July 3, 2013 - I've been a believer
in the value of search metrics as an advance market prediction methodology
within a focused reader context since I accidentally discovered the
correlations in my enterprise buyers guides in 1997 - and - more recently in
the SSD context I looked back on the past successes of the top SSD company
series in the Q1 2013.
I
was therefore amused to see that Google has recently published its findings
which report on the correlation of consumers searching and following up
information about new movies and how closely that aligns with later box office
revenues - which I read about in
a
recent article in Business Insider.
Choosing your next SSD
supplier is a bit more complicated than choosing which movie to watch - but the
principles are about the same.
People don't buy products and services
they don't know anything about - and search and learning about stuff indicates
possible intentions and often precedes serious actions.
Narrative is
important too. Maybe the movie sucks. Maybe the SSD company is doing sucky
things too. That's where I come in with some comments and some context. It's
not just about the numbers in the rankings.
I also have the benefit of
learning by reading reader emails. (But unlike some other agencies which have
recently been in the news - I only learn about stuff by reading and analyzing
the emails which have actually been addressed to me. Except when I've
accidentally block deleted them.) | | |
|
. |
|
| "Across the whole
enterprise - a single petabyte of SSD with new software could replace 10 to
50 petabytes of raw legacy HDD storage and still enable all the apps to run
much faster while being hosted on a shrunken population of SSD enhanced
servers." |
| the enterprise SSD
software event horizon | | |
|
.
. |
|
Reliability is more than
just MTBF... and unlike Quality - it's not free.
The battle for
storage reliability never stops.
It has to be fought - in every place
where physics intrudes on data integrity.
It must be fought and won
anew - in every technology generation and in every new product design. |
storage reliability
high availability
enterprise SSDs | | | |
