Founded in 2003, Diablo is successfully delivering products that enhance
the performance and capability of memory system designs.
Since 2010 - Diablo has been developing an
SSD technology which the company calls Memory Channel Storage which
leverages the disruptive capability of NAND-flash and future
technologies to enable increased levels of application performance supported by
a new class of enterprise server and storage system designs.
The Diablo executive management team, now developing its 3rd family of
memory interface solutions, has decades of experience in system architecture,
chip-set design and software development at companies including Nortel Networks,
Matrox Graphics, Goal Semiconductor, BroadTel Communications, ENQ Semiconductor,
IceFyre Semiconductor, Mosaid
Techologies and Huawei.
- editor mentions on StorageSearch.com
- editor's comments:- November 2013 -
Diablo Technologies has
created a new reference architecture for
enterprise SSDs -
which enables flash memory
based SSDs to work in conventional DIMM sockets which were originally designed
for DRAM - thereby
enabling apps speeds and latencies comparable to the
fastest PCIe SSDs
- but with a different software paradigm - where the flash aims to
transparently emulate DRAM style programmatic access instead of enforcing a
storage-centric view for APIs.
The first products based on this new
concept have recently been sampled to customers by Diablo's flash technology
collaborator (and route to market)
SMART Storage Systems
- which was recently acquired
articles related to
Diablo's place in SSD market
the fastest SSDs
the Top 20 SSD Companies
SSD glue chips and interfaces
SSD silos in the
Channel Storage / ULL SSDs
In November 2012
- Diablo Technologies
it has closed a $28 million
funding round which
will help the company
Memory Channel Storage concept to market.
In March 2013 -
announced it has closed an additional $7.5 million of funding, increasing
the total equity investment of its most recent round to $36 million.
- Diablo Technologies
SMART Storage as
its exclusive flash partner to pioneer a new type of (faster than
PCIe SSD) memory
In August 2013 -
it had begun sampling the first memory channel SSDs compatible with the
interface and reference architecture created by Diablo Technologies.
flash in enterprise SSDs
enterprise SSD users want?
11 key symmetries in
big market impact of SSD dark matter
thrive guide to enterprise SSDs
Auto-tuning SSD Accelerated
Pools of storage
don't all PCIe SSDs
look pretty much the same?
Efficiency as internecine
SSD competitive advantage
|more money for Diablo|
|Editor:- March 19, 2013 - It's
not unusual in the current
SSD market for
some companies to have oversubscribed
rounds - and that was the case recently with Diablo Technologies
it has closed an additional $7.5 million of funding, increasing the total
equity investment of its most recent round to $36 million.
previously announced investors.|
Editor's comments:- Diablo
describes its product as being these categories:- "soon-to-be-announced"
If you're already interested in
InfiniBand or SSDs
in RAM module form factors
(but which unlike the flash DIMMs from memory makers - might actually emerge
from stealth mode attached with a software model) then what Diablo does do
with its Memory Channel Storage may impact some of your future long range
Like you - I'm guessing of course. Diablo hasn't ordered any SSD
ads yet - although that $7.5 million would come in handy.
|Diablo names SMART Storage
as exclusive flash partner to pioneer memory channel SSDs|
|Editor:- April 25, 2013 - You may remember
reading here before about a company called Diablo Technologies -
which while in stealth mode - hinted it was working on a new technology which
would enable SSDs to run on server motherboards with latency and throughput even
better than PCIe SSDs....
Diablo has been creating the interface side of things - but I learned
recently that implementing the flash side of this - in a manner which is both
effective and affordable - requires a world leading mastery of enterprise flash
IP - which Diablo wisely recognized it doesn't have.
has publicly announced an exclusive partnership agreement with SMART Storage Systems
which will leverage its flash IP and controller assets to co-design a new family
of ultra-low latency SSDs and system accelerators which connect via Diablo's
memory channel storage architecture and which will be sold exclusively by
SMART but jointly supported by both companies.
comments:- there are a lot of implications for the future direction of SSD
server acceleration if this collaboration succeeds in delivering competitively
attractive new types of SSDs. But there are also very difficult technical
problems and ecosystems development problems to solve too in order to make it
I discussed these topics in a conversation earlier this week
with John Scaramuzzo,
President and Esther Spanjer,
Director SSD Marketing at SMART.
Among the many questions inspired
by that conversation:-
- how is the new technology different to what has been done before? -
particularly with PCIe SSDs and with DIMM class flash?
- if successful - what impact would memory channel SSDs have on the PCIe SSD
market? - and application server architecture?
- how will the new types of SSDs stretch the demands of flash endurance and
- how will competitors respond to this new technology? And how much of what
they say should you take take on board or disregard?
- who are going to be the among the first wave of customers to adopt these
I've written about these
matters and more in a new home page blog on StorageSearch.com -
Storage SSDs - will the new ultra low latency SSD concept fly? - should you book
a seat yet?
- when will the first products be ready?
up advisory team for new SSD interface|
|"As we prepare to launch our line of Memory
Channel Storage products that enable next-generation enterprise server and
storage system designs, we have set our sights on unprecedented levels of
performance for current and future applications To that end, we have
a group of top industry innovators to help refine the development of our
revolutionary NAND-flash system solutions... |
Maher Amer, CTO
January 29, 2013
|New to me
in SSD - Diablo Technologies|
|Editor:- November 8, 2012 - every week I learn
about new SSD companies and today one of them was Diablo Technologies -
which is working a new high speed direct attach like memory nand flash (and
maybe other nvm) SSD controller / product concept which it calls
it has closed a $28 million funding round which will help the company bring
their concept to market.
With this equity funding we are accelerating
the completion of a memory channel-based solid-state storage platform that will
deliver breakthroughs in system performance and flash storage density for
analytic data processing, web-page serving, cloud computing and other
server-based enterprise computing applications said Diablo's CEO and founder -
background is in chips (ASIC+FPGA), telecomms and DRAM memory. I haven't
spoken to anyone in the company yet.
But if I had to make a wild
guess it would be that Diablo's product may be something like
other PCIe SSD in
architecture - but packaged more like
compatible flash SSDs in form factor and connectivity.
note:- "What Diablo is doing is effectively creating a new layer in
the memory/storage hierarchy" said one of Diablo's investors Alex Benik in his
today explaining why he likes the company. "Diablos MCS-based products
will be dramatically higher capacity than DRAM, persistent, and deliver more
IOPS at lower latencies than any other SSD technology."
|memory channel SSD vs PCIe
SSD write latency - initial 3rd party benchmarks|
|Editor:- November 7, 2013 - Although it's
possible to make estimates of how a new type of
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
symmetries - it's only when you start to see
that you can have more confidence in your
models of what's
So if you've been wondering how Diablo's
storage compares with PCIe
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.
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
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.|
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
My best guess back in
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
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.
|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. |
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
interview with Diablo's flash partner SMART. So it was natural to ask how
Diablo viewed the impending acquisition of
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
- 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
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
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
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
- Is it bootable? No. Not yet that wasn't regarded as a priority.
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.
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.
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
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.
|McObject shows in-memory
database resilience in NVDIMM|
|Editor:- October 9, 2013 - what happens if you
pull out the power plug during intensive in-memory database transactions?
For those who don't want to rely on batteries - but who also need ultimate
speed - this is more than just an academic question.|
Recently on these
pages I've been talking a lot about a new type of
SSDs which are hoping to break into the application space owned by
PCIe SSDs. But another
solution in this area has always been DRAM with power fail features which save
data to flash in the event of
loss. (The only disadvantages being that the memory density and cost are
constrained by the nature of DRAM.)
products include in-memory database software) yesterday
published the results of
benchmarks using AGIGA
Tech's NVDIMM in which
they did some unthinkable things which you would never wish to try out for
yourself - like rebooting the server while it was running... The result?
Everything was OK.
"The idea that there must be a tradeoff
between performance and persistence/durability has become so ingrained in the
database field that it is rarely questioned. This test shows that mission
critical applications needn't accept latency as the price for recoverability.
Developers working in a variety of application categories will view this as a
breakthrough" said Steve Graves,
Here's a quote from the whitepaper -
Persistence, Without The Performance Penalty (pdf) - "In these tests
eXtremeDB's inserts and updates with AGIGA's NVDIMM for main memory storage
were 2x as fast as using the same IMDS with transaction logging, and
approximately 5x faster for database updates (and this with the
transaction log stored on RAM-disk, a solution that is (even) faster than
storing the log on an SSD).
The possibility of gaining so much speed
while giving up nothing in terms of data durability or recoverability makes the
IMDS with NVDIMM combination impossible to ignore in many application
categories, including capital markets, telecom/networking, aerospace and
Editor's comments:- last year
McObject published a paper showing the benefits of using PCIe SSDs for the
transaction log too. They seem to have all angles covered for mission critical
ultrafast databases that can be squeezed into memory.