 |
| leading the way to the
new storage frontier | |
.. |
| ..... |
|
the Top SSD Companies
an SSD
conversation with PLX
boom bust
cycles in memory markets
meet Ken and the SSD
event horizon
how
fast can your SSD run backwards?
optimizing
CPUs for use with SSDs in the Memoryfication Era |
|
| . |
|
 |
|
. |
|
| SanDisk's new software
platform repurposes SSDs as storage class memory |
Editor:- July 8, 2014 - 2 weeks ago
SanDisk
announced
a new enterprise software product -
ZetaScale
(pdf) - designed to support large inmemory intensice applications.
I
delayed writing about it at the time - until I learned more. But now I think it
could be one of the most significant SSD software products launched in
2014
- because of the freedom it will give big memory customers (in the next 2-3
years) about how they navigate their tactical choices of populating their
apps servers with low latency flash SSD hardware.
what is
ZetaScale?
SanDisk says - "ZetaScale software's highly
parallelized code supports high throughput for flash I/O, even for small
objects, and optimizes the use of CPU cores, DRAM, and flash to maximize
application throughput. Applications that have been flash-optimized through the
use of ZetaScale can achieve performance levels close to in-memory DRAM
performance."
ZetaScale is SSD agnostic. "ZetaScale is
compatible with any brand of PCIe, SAS, SATA, DIMM or NVMe connected flash
storage device, providing customers the ability to choose, avoiding hardware
vendor lock-in."
I was curious to see how this new product -
which is a toolkit for deploying flash with tiering to DRAM as a new memory
type - fitted in with other products - from SanDisk and from other vendors which
also operate in this "flash as a big memoryalternative to DRAM"
application space .
So I asked SanDisk some questions - and got some
interesting answers.
- Where does the ZetaScale product come from?
SanDisk -
ZetaScale builds upon our
Schooner
acquisition technology for additional use cases and flash deployment models.
ZetaScale allows any developer to better tune their applications for
flash-based environments, no matter which vendors hardware or interface is being
leveraged. Thus, ZetaScale represents a major step forward in our vision of the
flash-transformed data centerempowering software developers to scale and enhance
their applications to meet today's big data and real-time analytics demands,
while lowering TCO.
- How much commonality is there between ZetaScale and FlashSoft product
offerings?
ZetaScale and FlashSoft software are complementary and
orthogonal.
FlashSoft provides direct-attached flash-based
caching for NAS and SAN devices, with the goal of improving performance for
unmodified applications running on a server.
ZetaScale
software provides a flash and multi-core optimization library that applications
can integrate to allow them to achieve 3x times the performance
improvement from flash alone.
Both ZetaScale and FlashSoft software
provide their benefits in bare metal and virtualized environments
- Does ZetaScale support ULLtraDIMM?
Yes. The software is compatible with any brand of PCIe, SAS,
SATA, DIMM or NVMe connected flash device, enabling users to avoid vendor
lock-in. However, the software does not get embedded into any SSD.
- How would ZetaScale fit into a future SanDisk product line which also
includes Fusion-io?
SanDisk cannot comment on open M&A activity. As usual, all
planning surrounding the product portfolio and roadmap will begin following the
close of the acquisition.
Editor's comments:- overall I'd have to rate SanDisk's -
ZetaScale as one of the most significant SSD software products launched in 2014.
From a technical point of view - it's a toolkit which will enable
architects of SSD apps servers with very large in memory databases to
decouple themselves fromdeep dives into specific low latency SSD products.
Instead of gambling on whether they should exploit particular features which
come with particular low latency SSDs - they can instead use ZetaScale as the
lowest level of flash which their apps talk to. And that will change markets.
And
although SanDisk didn't want to comment on how this would be positioned against
Fusion-io's VSL - it's
undeniable that in some applications it does compete today.
Although I
wouldn't be surprised to see - a year after the acquisition (if it goes ahead)
ZetaScale could be useful as a way of introducing new customers to the
ioMemory hardware environment - without those customers having to make a hard
commitment to the rest of Fusion-io's software.
And - looking at the
memory channel
SSD market - it also means that SanDisk software might be a safer standard
for future customers of any DDR4 or HMC SSDs which might emerge from competitor
Micron which - unlike
SanDisk - hasn't demonstrated yet any strong ambitions in the
SSD software platform
market.
Later:- A year after writing the above comments
Micron gave the first
public sign that it was interested in tackling the storage class memory market -
with its
3D
Xpoint announcement.
However with products not expected till 2016
- it confirmed my assessment that Micron was at least 2 years behind the market
curve on a usable SCM platform at the time of the ZetaScale launch.
What
do want our memory systems to do for us?
And how do we want them to
behave?
Those questions were tackled in a later blog -
where are we
heading with memory intensive systems and software? | | |
|
. |
|
| We used to be able to write
down a list of memories which mattered. Now we're seeing many new strangers
from strange lands. Even I can't remember who they all are and how to spell
their names |
| the SSD
Bookmarks - series 2.0 | | |
|
. |
|
 |
|
. |
|
| Decloaking hidden segments
in the enterprise |
Editor:- StorageSearch.com recently
published a new article -
Decloaking
hidden segments in the enterprise for rackmount SSDs
Some of the
world's leading SSD marketers have confided in me they know from
their own customer anecdotes that there are many segments for enterprise
flash arrays which aren't listed or even hinted at in standard models of
the enterprise market.
Many of these missing market segments don't
even have names.
Hey - that means SSD-world is like a map of the
US before Lewis and Clark.
If you're a
VC should this make
you anxious or happy?
If you're a user - maybe that's why no one is
delighting you in the way you think you deserve.
That's what led me to
write my new article.
See
also:- rackmount
SSDs, SSD silos,
market research | | |
|
. |
|
|
|
|
. |
|
|
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
| Although the many stories
about the true nature of the stars, moon and other heavenly bodies in the night
sky were created and developed over many thousands of years - the SSD market -
which has a much shorter history (spanning about 40 years) - has nevertheless
managed to accrue an imaginative body of literature which includes truths, half
truths, mysticism, misunderstandings. myths, legends - and in some cases -
downright balderdash - when it comes to the subject of SSD costs, pricing and
justifications. |
| Exiting the
Astrological Age of Enterprise SSD Pricing | | |
|
. |
|
|
|
. |
|
|
|
|
. |
|
| stories
from SSD news in July months gone by |
| July 2013 |
Diablo unveils Memory Channel Storage |
| July 2012 |
LSI ships 1 million SandForce controllers / month |
| July 2011 |
8 out of top 10 SSD companies support PCIe |
| July 2010 |
SNIA publishes draft SSD performance testing doc |
| July 2009 |
STEC announces $150 million of SSD orders |
| July 2008 |
Samsung and Sun work on higher endurance SSDs |
| July 2006 |
HyperStor SSD systems scale to Tens of
Terabytes |
| July 2004 |
Adtron introduces first industrial SATA SSD | | |
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
| So you'd think that when it
comes to the person in purchasing - who's looking for standard SSD products to
go into industrial or embedded systems - which aren't pushing the state of the
art when it comes to form factors (some like CF have been around forever) and
whose needs include the type of SSD interfaces which haven't been seen much in
SSD news in the last 10 years - it should be simple enough. |
| no such thing as a simple
standard legacy industrial SSD - (June 3, 2014) | | |
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
|
|
|
. |
|
|
|
..... | |
|
|
| .. |
 |
| .. |
|
|
|
Diablo secures patent
related to MCS technologies
Editor:- July 31, 2014 - Diablo today
announced
that the US Patent and Trademark Office (USPTO) awarded U.S. patent No.
8,713,379, entitled a "System and method of interfacing co-processors and
input/output devices via a main memory system," to the company. The '379
patent describes:
- A method for connecting non-volatile memory directly to the memory
controllers of a processor.
- A learning machine to handle data interleaving/de-interleaving and data
scrambling/de-scrambling algorithms for DDR3/4-based memory
controllers.
- A method to remap the non-linear DIMM address space back to linear address
space used by the driver
See also:-
Memory Channel
SSDs
Lite-On enters enterprise M.2 PCIe SSD market
Editor:-
July 30, 2014 - Lite-On
will be showing a new M.2 PCIe
SSD for the enterprise market next week at the Flash Memory Summit.
The P1P is a
M.2
PCIe SSD with capacities of 1TB and
power loss
protection circuit all in a small form factor.
The small form
factor and high capacity allow enterprise customers to pack more storage in a
smaller footprint. The P1P Series can deliver sequential R/W speeds up to
610MB/s and 520MB/s while random R/W speeds can be upto 95K/15K
IOPS.
The
drive has a MTBF
of 2 million hours and an
endurance
rating of up to 1 drive write
per day for 5 years.
See also:- who's who
in M.2 SSDs?
NxGn Data exits stealth with promise of in-situ SSD processing
Editor:-
July 29, 2014 -
NxGn Data today
exited stealth mode.
NxGn will use
advanced
adaptive DSP technology to enable small form factor SSDs (such as
M.2)
aimed at the enterprise market - using MLC and TLC down to 1z-nm geometries.
Fully functional FPGA-based samples will be available in early 2015,
followed by final production samples of SoC-based M.2 solutions in late 2015.
NxGn
says it will be the first SSD controller company in the industry with
in-storage computation capability - what it calls "In-Situ Processing".
Editor's
comments:- Earlier this year I published a couple of reports and mentions
about SSD suppliers (LSI
and Memblaze) who
have modifed their controller firmware to eliminate or bypass functions from the
lowest level SSD drive - for large customers like Baidu - who then use their
array level software to get better utilization and performance.
And in
the industrial
market InnoDisk - uses
what it calls 3rd generation architecture - to partition intelligent data
actions between the controller and software stack.
Until recently -
only Fusion-io (in
whose products the flash controller and apps server - share the same CPU cores)
has been able to maximize high end context intelligence with low level flash
block data access at a similar latency level.
But once you've solved
the problem of making SSDs reliable and fast - it's tempting to create an SSD
instruction set which which focuses on application layer needs too - and not
just those of dumb storage.
See also:-
Active
Flash: Towards Energy-Efficient, In-Situ Data Analytics on Extreme-Scale
Machines
real-world performance of flash storage systems
Editor:-
July 23, 2014 - Editor:- July 23, 2014 - How does flash storage perform in the
real world? - Demartek
aims to provide some answers by reporting on the performance tests
which it has carried out on SSD and hybrid systems from many of the leading
enterprise SSD companies in a session next month at the Flash Memory Summit (August
5).
Demartek says attendees will come away with reasonable estimates
of what they can expect in practice and the results also reveal additional
advantages of flash-based storage, with what Dennis Martin, President
- Demartek calls "happy side effects". ...more
info (pdf)
See also:-
SSD testing & analyzer
news, how fast
can your SSD run backwards?
STT-MRAM? - update report
Editor:- July 18, 2014 -
IEEE Spectrum today
published an interesting state of the art article about spin-transfer-torque
MRAM -
Spin
Memory Shows its Might.
Among other things the article's author -
Rachel Courtland
Editor-in-Chief of IEEE Spectrum Magazine - says "STT-MRAM may be claiming
some of the enthusiasm once reserved for other alternative memories, such as
ferroelectric RAM, phase-change memory, and resistive RAM. But its success will
come down to manufacturing technology and how well it can compete on cost."
...read
the article
See also:-
flash and other nvm,
storage interface chips
new article on enterprise SSD pricing
Editor:- July
18, 2014 - in a new article on StorageSearch.com
-
Exiting
the Astrological Age of Enterprise SSD Pricing - I explain why I think that
2014 will be seen as the start of a new phase of creativity in the enterprise
SSD market on the subject of pricing and affordability and I name 3 companies
leading this charge.
It shouldn't come as any surprise that the
dominant form factor leading this new market trend is
rackmount SSDs.
because as I told you in this article last year -
exciting new
directions in rackmount SSDs - that's the most productive form factor for
doing something efficiently
different with technology.
And the simplest way for vendors to signal
to the world that they are masters and commanders of the solid state storage
high seas - rather than merely floating barges of nand flash which can be
swept along in any direction by the latest technology gust - is to hoist new
colors of SSD pricing. ...read the
article.
SSD brand leaders - from IT Brand Pulse
Editor:-
July 17, 2014 - Based on votes by end users in its survey sample groups -
IT Brand Pulse
today
named
the "2014 SSD Brand Leaders".
The IT Brand Pulse "Innovation
Leaders" included these companies and categories:-
Editor's comments:- with so many companies now doing market
research which intersects with the SSD market - you've now got more sanity
checks than ever to determine whether your assessment of any particular SSD
company is broadly in line with that of other people.
Brand
awareness, financial reports and online search data - intrinsically provide
different numerically weighted views of the same market. And which methodology
you prefer depends on whether your priority is in understanding the past,
present or future - and what it is you're trying to decide.
For
longitudinal market studies you can refine your understanding of shifting
patterns of market change and real leadership by comparing these different
types of data and correlating the movements in different time periods. So a
shift in search volume in one period may correlate to a change in revenue some
time later. Or a change in revenue or profitability might be seen to be tied to
a change in brand strength later.
The interesting thing about the SSD
market is that because the technology hasn't been standing still - it can be
shocking to compare the same list and see how it changes over a period of 3 to
5 years and to see which old names have disappeared and which new ones have
replaced them. I'm often reminded of this when I trawl back through my own
SSD history
and news archives. I'm sure it's the same for many of you too.
One of
the things we all hope to get out of these lists - is to avoid making too many
bad bets on companies and technologies which will prove to be a waste of our
time.
the Top SSD Companies
storage market research
Can you
trust SSD market data?
Tegile appoints former Violin VP as new CMO
Editor:-
July 15, 2014 - Tegile
Systems today
announced
that Narayan
Venkat - who was formerly VP products and marketing at Violin - has joined
Tegile as its Chief Marketing Officer.
See also:-
rackmount SSDs,
playing the
enterprise SSD box riddle game
OCZ announces availability of ZD-XL SQL Accelerator 1.5
Editor:-
July 15, 2014 - OCZ
this week announced availability of
version
1.5 of its ZD-XL SQL Accelerator (PCIe SSD and caching software bundle) the
beta version of which was announced in
April.
Highlights:-
- flash Buffer Pool Extension (BPE) support
- better granularity of database files that need to be accelerated - than
version 1.0.
- remote flash services - enables remote network connected blade servers
to access cached PCIe flash storage using OCZ's proprietary Direct Pass Caching
Technology.
Small system performance? - Internal testing performed by OCZ
delivered over one million TPMs for a sample of 50 virtual users when ZD-XL SQL
Accelerator 1.5 was located in the same server as the SQL application.
For
larger enterprises/data centers that use ZD-XL SQL Accelerator 1.5 in a
traditional HA SAN storage environment, the embedded software can cache large
database files from the SAN onto server-side flash either locally or remotely -
with 5x speedups (compared to native HDD SAN performance) being
realistic goals.
Editor's comments:- This product is aimed at
the same applications as Violin's
Windows
Flash Array (WFA)
The main differences are:-
- OCZ has been selling into the entry level enterprise SQL
acceleration market as a business focus for much longer than Violin (years
rather than months).
In contrast - Violin's history has been mostly as
a very high performance SSD supplier - and until recently Violin only
encountered small SSD applications as departmental use cases within big SSD
customer sites.
- With Violin's WFA you get an integrated system - but have to buy the
servers from Violin.
- With OCZ's ZD-XL SQL Accelerator - you have the freedom to use any servers
you like - as long as they have enough spare slots to install the PCIe SSD
cards. But you have to know a bit more about what you're doing - and the
performance is unique to your system.
See also:-
ZD-XL
SQL Accelerator 1.5 case studies and white papers
an update on Atlantis (Software-Defined Storage)
Editor:-
July 14, 2014 - Atlantis
Computing today
announced
2 things:-
- On a year-over-year basis Atlantis grew 1H bookings 80%
Commenting
on the state of the company he founded in 2006 and which has already sold over
580,000 licenses - CTO, Chetan Venkatesh
said - "Atlantis was started with a vision of a future where intelligent
and agile Software solved the most important, complex and challenging problems
of storage hardware. In our view, storage is not a hardware problem, but that
hardware is the problem and to solve that you need intelligent software that
disrupts the current paradigm of expensive and proprietary SAN & NAS."
Avere gets another $20 million funding
Editor:- July
10, 2014 - Avere
Systems today
announced
raised an additional $20 million in
venture financing,
bringing the total amount invested in the company to $72 million. The Series D
funding round was led by Western Digital Capital, with participation from
previous investors Lightspeed Venture Partners, Menlo Ventures, Norwest Venture
Partners and Tenaya Capital. The funding will be used to accelerate sales,
marketing and continued development of the company's hybrid cloud storage
solutions.
"The reality for 99% of enterprises is they
will operate increasingly in a hybrid IT storage environment for many years to
come. This means that no single storage technology will win, and both
on-premises and cloud storage will be required to achieve cost and performance
goals" said Ron Bianchini,
president and CEO of Avere Systems.
Editor's comments:- I
agree with Ron about that 99% figure.
In my 2012 article -
an introduction to
enterprise SSD silos - when I was writing about a future in which SSDs are
everywhere - I said - "No single SSD type can match all the needs of all
user enterprises economically. And there will always be a need to have
intermediate management between SSD systems which have dissimilar speed / cost
characteristics." - "SSD systems" in this context - includes
the cloud.
But
I'd like to take Ron's statement - about what constitutes a "hybrid"
- a bit further...
I would add the words "software
architectures" somewhere in that strong assertion about hybrids. (Although
I can see why that would be too messy to include in a press release.)
I think that enterprises today - whether they consciously realize it or
not - are in reality choosing between at least 4 different generations of
SSD-aware software architectures within and around their enterprise SSD
hardware mix - every time they buy a new storage product or server - in
addition to the straightforward and highly visible determinations they make
about the 3 fundamentally different types of
hybrid caching and
tiering SSD appliances (different from a network architecture point of
view) which we've seen operating in classic
legacy storage
software frameworks since 2009.
Those "software hybrids"
- which represent different generations of thinking and different pragmatic
business approaches - are at the root of the software based multipliers in
rackmount
SSD market segments (which are making everyone's life more complicated).
In
the long term the impact of multi-generationally-rooted and SSD-centric
software hybridization in the enterprise will be as significant as the
differences between a
2.5" SSD and a 2.5"
hard drive.
I
have written about these software generations before in various articles - so I
won't repeat them here - I might write a simple bullet point unified history
of this subject if enough readers ask me to clarify this.
a reader asks - about continuity of PLX's PCI Express box
platform
Editor:- July 10, 2014 - Yesterday a reader asked if I
thought that the PCIe fabric system - which I had written about last month in
the article -
An SSD
conversation with PLX - would still be available as a product after the
acquisition of PLX
by Avago Technologies closed?
He
pointed out that Avago avoids competing with its own systems customers (which is
why they divested the SSD business from LSI) and also Avago isn't a systems
company.
Here's what I said...
The points you make are valid
and I had wondered the same thing myself.
But I think the PCIe fabric box (ExpressFabric solution)
will continue in some form or other – because it started out as a system
design kit for demonstrating what could be done with the next generation of PCIe
chips and software stacks. So it's an essential sales tool for the chip
business.
Before the announcement from Avago – the PLX guys had already
said that any of their customers would be welcome to use as much or as little as
they wanted from the SDK box – as PLX wasn't originally advancing this as a
systems product.
But I could see from my own judgement (and PLX
confirmed this) that for some customers – having availability of such a
box as a product would be a convenient tool.
Incidentally - that's
how Intel got into the systems business - with their
Multibus
SBC product range in the 1970s - which was a response to customers asking
if they could buy the boards which were in the early microprocessor development
systems.
Returning to Avago and the PCIe fabric SDK
I agree -
it's possible that Avago may decide not to get into the volume supply
business of these boxes – but in that case I think the boxes (and design
IP for these boxes) would continue to be available in some form from a
designated source – otherwise lack of this integration concept tool would
slow down market adoption of the new PCIe fabric chips.
It's in the
interests of Avago and its oem customers to ensure that the ExpressFabric SDK
remains available as a software reference architecture which is at the center
of this new ecosystem.
Later:- I asked for an official response about
this (just in case I had missed anything) but - as I expected - the answer
was - it will come in a future press release.
EMC acquires cloud piped via iSCSI company TwinStrata
Editor:-
July 8, 2014 - not that it really matters - EMC
has acquired TwinStrata | |
| . |
 |
| . |
|
| |
| .. |
 |
Michelangelo was looking for David.
Megabyte
was looking for the inner SSD. |
| |
|
.. |
|
|
|
. |
|
|
|
. |
|
Can you trust market
reports and the handed down wisdom from analysts, bloggers and so-called "industry
experts" any more than you can trust SSD benchmarks to tell you which
product is best?
The short answer is - heck no! - whatever gave you that silly idea?
|
| Can you
trust SSD market data? - the classic article | | | |
