click to visit StorageSearch.com home page
leading the way to the new storage frontier
...
.....
after AFAs - what's the next box?
how fast can your SSD run backwards?
capacitor hold up times in 2.5" military SSDs
where are we heading with memory intensive systems?
optimizing CPUs for use in SSDs in the Post Modernist Era
controllernomics and risk reward with big memory "flash as RAM"

...
SSD news highlights in May 2017
by Zsolt Kerekes, editor - StorageSearch.com
Thinking. That's the hard part about the SSD market. And there's a lot for you to think about in a couple of recent stories on this SSD news page.
  • MRAM? - should we still be calling this - "emerging memory"?

    In my experience a key dividing line for new technologies is when their creators put money into setting up sales channels and advertising.

    Because although you can sell visions to investors the real test is selling tangible products to customers.

    Everspin recently set up an enterprise sales channel, and this month started talking about how many of its MRAM chips are going into U.2 SSDs.
  • Western Digital has paused the sale of Toshiba's memory business.

    Delay is good for WDC but bad for Toshiba and other potential buyers.

    The tensions are analyzed in recent news stories.

    Would a WDC acquisition be good or bad for the storage market?
  • 3D, 3D, 3D...

    Everyone is talking about their latest 3D nand achievements (memory, SSD, array) and how they are solving the world's problems.

    You may learn a lot from a simple question posed by a semiconductor memory company which competes with nand...

    Is Micron's 3D memory (really) cheaper than 2D?

    In some ways this is an old story which goes right back to the dawn of the modern SSD era (the period since when products external to the SSD ecosystem no longer triggered extinction points for SSD roadmaps).

    In past classic articles I have discussed the differences between SSD user value propositions and SSD price.

    But I despite all that I anticipate you will learn a lot about the thinking in the semiconductor memory industry by following the threads above.

    It's a casino in which SSDs tend to win more often than the memories in my view.

    This didn't matter when the SSD market was small - but as I wrote in 2013 - everyone's now hostage to the fortunes of SSD.

    Even Seagate - which in April 2017 - tried to minimize the issue by saying that only 10% of its HDD products were now at the risk of future substitution by SSDs.

    That's an old argument too.

    And I will return to it with a new analysis of what the extinction threat really may be for all those future hard drives which have relocated themselves in the safe market havens of video surveillance and cold storage.
...
SSD ad - click for more info
...
Micron dares skin in the SSD box game
Editor:- May 5, 2017 - Micron this week announced its long overdue plans for entering the rackmount SSD market.

Micron's ambitions revolve around SolidScale - a 2U box stuffed with NVMe SSDs and interconnected in a fabric using software from Excelero.

Micron is seeking early validation customers and technology partners now and says its platform is expected to begin volume production in early 2018.
SSD news May 2017 - Micron's new box
Editor's comments:- This is a welcome - albeit long overdue - strategic move by Micron.

Two years ago (in June 2015) I commented on Micron's weak understanding of the enterprise SSD systems market and I said this...

"Micron is the only company (among the major memory and HDD companies in the same market) which hasn't acquired any significant enterprise SSD companies in the past 12 to 18 months. For years they didn't understand the enterprise and they still don't have the IP or brands needed. So they should feel strong pressure to do something. But making a big move after years of doing nothing is risky too."

The core of my (2015) downgrade on Micron's enterprise business was:-

It's only possible to deliver the most efficient (and therefore competive) SSD based products at the box level due to the accumulated incremental advantages of big controller SSD architecture (managing large numbers of flash devices) compared to optimizations made at the single drive level.

And the box level is the lowest level at which multi-layered and multi-location intelligence can be usefully managed by internal software which is applications and box aware and not just aware of tricks at the solo flash drive level.

Furthermore it was clear that the enterprise market will change due to growing customer experience and expectations leading to a much smaller set of systems level products supplying the whole market.

In this predicted vein SanDisk's Infiniflash - launched in March 2015 - was one of the early examples of a standard commodity SSD box being offered by one of the big semiconductor memory or HDD makers.

Although you can glean useful market lessons from the adoption (or not) of Infiniflash (which was a box of SAS SSDs) and try to apply them them a prediction of the prospects for SolidScale - I think the most useful lessons are at the tactical rather than strategic level.

As you'd expect from any bunch of SSDs in a box - Micron's SolidScale is currently lightweight and lacking any proprietary depth in box scale IP.

From the technical point of view - there is nothing that SolidScale can do which can't be done equally well and sooner by packing similar SSDs into another branded box and using the same or similar externally sourced software.

On the plus side - for Micron - it will learn useful market lessons by having some skin in the SSD box game. SSD history shows that it takes many years before players are taken seriously by the enterprise market (typically because of the need to integrate HA features and software) - but new box vendors can only get better by playing or acquiring better players.

Customers in more technically adventurous markets like web entities, cloud infrastructure and HPC have been the most receptive early adopters of performance based drives (PCIe SSDs) and fast boxes. And that's a direction to which Micron can turn to get early customer feedback.

On the downplaying expectations side:- for Micron...
  • the PCIe fabric box is not the whole networked storage box market
  • the NVMe SSD array market will itself become fragmented into many latency zones
  • the NVMe SSD array market (latency optimized storage) will not be the whole enterprise SSD storage market.
  • the use of memory systems architectures is changing and will likely evolve to downplay the relative size of local storage compared to memory - towards cloud adapted memory systems
Segmentation challenges in the enterprise memory systems market have been complex in past years, leaving many viable user needs unmet by satisfactory products and the disruptive interplay of CPUs, memory and software in the Post Modernist SSD Era means that no current vendor business plans or roadmaps should be regarded as being sacrosanct or endurable.

Having said all that...

It's better for Micron to have a box business with which it can play and learn than merely supplying the turf to other big box (stadium) brand leaders.

Who dares wins (if they survive).

Here's a short timeline of related articles.
  • new SSD thinking inside the box (2013) - some of the deepest thinking going on in the SSD market is focused on rackmount SSDs, but there will continue to be diversity in the internal architectures and detailed features due to valid different preferences among end-users
  • playing the enterprise SSD box riddle game (2013) - for many enterprise Users - the question - can you guess what will my next SSD box will look like? - bears a striking resemblance to the unfairness of Gollum - when he says - what has it got in its pocketses - Precious?
  • Decloaking hidden rackmount SSD segments in the enterprise (2014) - A contributory cause for gaps in segmental understanding has been the continuing pace of disruptive innovation in enterprise SSDland - which has meant there hasn't been a stable market template for vendors to follow from one seemingly chaotic year to the next as they encroach on new markets.
  • Exiting the Astrological Age of Enterprise SSD Pricing (2014) - vendors and users face different sets of uncertainties and challenges - but the risk of not doing business is bigger than the inevitable mistakes which will be made - innovative pricing strategies can provide plausible calm assurance that everything is under control
  • we're better than EMC - the unreal positioning of many AFA startups (2015) - For investors in such startups - consider this. When your flash enterprise vendor spends so much energy talking about competitive comparisons in a market segment - which is so restricted (EMC legacy base) and about theoretical opportunities which aren't real and which can never scale upwards (because the big customers have spent more on EMC than your pet AFA company received in its last funding round) - then they're wasting their share of voice in the market. .
  • after AFAs - what's next? (2017) - Every assumption which served you well about the relative sizes of memory and storage in your enterprise world will soon be wrong. Even "AFAs everywhere" are a transient market stopover and no longer the final destination of the solid state storage roadmap.
...
SSD ad - click for more info
...
"Re the nand shortages... I think it's probably more challenging for the technology industry to deal with the shortages than it is maybe beneficial for HDDs because of some comparison on say some 500GB drive.... It's not easy to say if it's good or bad... We've always said it's a better world if there's a lot of nand because that means people have more devices in their hands and they're creating more data."
Stephen Luczo, Chairman and CEO - Seagate in Q3 FY17 conference call - April 26, 2017.

Editor's comments:- other interesting things which Seagate mentioned in this conference call...
  • the company thinks that only 10% of its current HDD portfolio is feasibly replaceable by flash.

    That's because the company has chosen to engage in the HDD market with higher capacity hard drives, and has for years been cultivating applications such as video surveillance (which don't need high random IOPS).
  • Seagate's flash and cloud systems business grew sales 19% yoy.

    The company's assessment is that its SSD sales could have been maybe upto $50 million higher had it not been for shortages of memory.
see also:- spinning down to HDD's retirement
...
SSD ad - click for more info
...
cool runnings

Rambus to coach faster DRAM
Editor:- April 20, 2017 - Back in the early 1990s it was not uncommon to hear about specialist server companies which were using peltier effect heat sinks to refrigerate the fastest workstation processors so that they could be run at higher clock speeds. But this kind of extreme approach to server acceleration only provided short term competitive gains in a single dimension.

One of the biggest bottlenecks in the past decade has been RAM architecture and DRAM implementation itself. (You can read more articles about the background to this on the DRAM resource page here on StorageSearch.com.)

A new angle on extending the performance of DRAM was announced recently by Rambus and Microsoft who are collaborating on the design of prototype super cooled DRAM systems to explore avenues of improvement in latency and density due to physics effects below -180 C.

A new article - Rambus, Microsoft Heat Up With Cold DRAM - by Junko Yoshida , Chief International Correspondent - EE Times - discusses these plan in more detail.

In the article - Craig Hampel, chief scientist at Rambus, told EE Times that "Microsoft isnt alone... heavy data center users like Google, Facebook and Amazon are all in search of new memory architecture. Indeed, these tech giants who have primarily grown their business via their technological prowess in software development are now finding the future of their business growth severely constrained by hardware advancements." ...read the article

Editor's comments:- At room temperature the main problem in fast DRAM systems is that the energy required for refresh cooks the chips which means cells lose charge faster which creates data integrity risks which in turn needs more frequent refresh.

This is a limiting design factor.

It means that even if you have a miraculous packaging technique which can sandwich more chips into a box - DRAM loses out to nvm technologies which don't require refresh - when the scale of the installed capacity (and watts) in the box is high.

Because if you can't fit enough RAM into the same single box then the memory system accrues a box-hopping fabric-latency penalty which outweighs the benefits of the faster raw memory chip access times inside the original box.

If you freeze DRAM then the refresh cycle can be extended (which means you can pack more capacity in a box) but also the native transit time for data in the copper interconnects and inside the silicon gets faster too.

Although Rambus and Microsoft are pitching this a progressive research exercise I don't think that it will provide a general solution for data intensive factories.

While it's a good thing for researchers to play around and explore the limits of what can be done with all kinds of memory devices - I think that the answer to greater performance lies in new architectures rather than freezing old ones.

...
Cadence logo - click for more info

sponsors StorageSearch.com

SSD news

earlier news / popular articles / SSD controllers

new edition of the Top SSD Companies

Editor:- May 19, 2017 - The Top SSD Companies in Q1 2017 - was published today by StorageSearch.com.

This is the 40th quarterly edition in this market defining series.

We're still seeing many signs of significant changes in the pipeline which will sweep away and redefine the relative importance of existing SSD technologies and product lines.

You might say - that's SSD business as usual.

And in some ways it is. There are few rafts of stability and consistent reference in this crazy market of ours. This series is one of them. These are the companies which mattered most to other readers like you in the recent period. ...read the article

PS - I accidentally had to make up a new word to describe some of the changes we're seeing - as in the "future memoryfication of the enterprise".

Sorry about that. If I find something better I might replace it. But for now it's the most apt word to describe what's going on.


new patents for Corsair and Violin

Editor:- May 18, 2017 - News about the issuance of new patents in the SSD and memory market this month include:-
  • Corsair - 9,645,619 - described by its inventor Bobby Kinstle Senior Project Manager in this way. "This one is for using tiny heat pipes to remove heat from memory devices in really tight spaces."
  • Violin (now under new ownership) announced 2 new patents.

    9,495,110 - technology behind the company's 4-way active-active RAID controllers. "Method for managing storage memory resources for LUN allocation and related maintenance operations within a system of high-availability distributed RAID controllers, while maintaining coherency and high input/output speed."

    9,547,588 - "Method for managing the compression and/or encoding of digital data being stored in non-volatile memory, such as flash memory, to improve the performance of the solid state memory cells with respect to program/erase operations. This method may reduce power consumption and storage space usage, while enhancing the wear lifetime characteristics of the memory media."



TrendFocus reports sequential decline in raw storage drive capacity sold in Q1 2017

Editor:- May 16, 2017 - TrendFocus today announced market highlights related to storage drives in Q1 2017.

The number of exabytes shipped in HDDs and SSDs fell sequentially by 5% but was 19% higher than the year ago period.

TrendFocus says several factors contributed to the decline:- including weak PC sales, cyclicality in data center spending, and tight NAND supply affecting SSD buying patterns.

Re delays in storage spending in data centers:- TrendFocus says...

"Some large hyperscale/data center companies purchased fewer HDDs and SSDs in the quarter, temporarily slowing exabyte growth in the segment responsible for driving the highest long-term rate of capacity demand."

Editor's comments:- whereas higher prices of flash SSDs due to fab production capacity constraints and lower than expected yields in new memories has undoubtedly been a significant factor in the market - another thing to keep your eye on is the potential for big users to get more use out of the same raw capacity by using newer software.

That's a factor which can surprise the market at any time as I discussed in my warning article about the impact of the SSD software event horizon.

Another long term trend which will depress storage drive sales will be the refocus on memoryfication architectures.

I'm not going to attribute this 5% decline in raw storage drive byte shipments to the adoption of new memory architectures. It's probably just a correction with many causes.

But looking ahead to the next several future years you should not be surprised to the impacts on the storage drive market which could be much higher.


Toshiba's memory business sale - tensions in the timing

Editor:- May 15, 2017 - progress updates about the sale of Toshiba's SSD and memory business have been part of the underlying news fabric during the first half of 2017. Tensions in the timing and conclusion of a prospective sales come from various causes:-
  • Toshiba needs the promise of a buyer as soon as possible to avert further downgrading of the parent company due to losses in other markets.

    An additional urgency is that the memory market was in a boom at the start of the spin-off process - with prices for memory rising.

    If the market changes back to oversupply or if other factors reduce the premium value placed on raw flash - before an acquisition is agreed - then the value of the business will be less.

    Delay therefore works to the advantage of a buyer and to the disadvantage of the seller.
  • The size and strategic nature of the transaction means that geo-political considerations and anti-trust hurdles will play big part in the decision process.

    If the potential buyer triggers a monopoly investigation in their own operating markets then delay in the acquisition is inevitable and could take upto a year.
  • A complicating factor is that Toshiba shares considerable flash wafer fab assets and IP with Western Digital which were part of arrangements evolved in partnership with SanDisk -which WDC acquired in 2015.

    In the current situation where Toshiba's memory business is for sale - there could be significant differences in self interest between WDC and any future owner of the Toshiba memory, SSD and HDD business.
In that context these reports about WDC's growing unease can be more easily appreciated:-
  • Nikkei Asian Review (April 21) - "Western Digital remains opposed to any sale of Toshiba's memory unit to a third party."
  • FT (May 15) - "Western Digital is taking its row with Toshiba to the International Court of Arbitration, as the US storage technology group tries to prevent what it claims is a breach of the terms of its joint venture with the embattled Japanese conglomerate."
Editor's comments:- It's clear that in the matter of who buys Toshiba's memory, SSD and HDD business the strategic and business self interests of WDC and Toshiba (parent company) are different.

But what would be best for the industry?

My own view is that if WDC succeeded in acquiring all of Toshiba's memory and SSD assets in 2017 or 2018 the effect would be to reduce the ability of the SSD market to be as competitive internally and also (although less important) reduce the ability of the SSD market to compete as aggressively with the hard drive market.

Another factor is the effect of competition inside the memory market and the effect on the emerging market for tiered big memory.

This is a market which is working to the benefit of users because there are pressures for flash to displace DRAM and a market climate which is nurturing the growth of new nvms which (although small in revenue terms now) could in the long term displace a fraction of the flash or DRAM markets.

In 5 years time such an acquisition (a major memory supplier being acquired by one of the two remaing hard drive companies) may not matter so much - because the SSD and memory systems market will have moved to new levels of efficiency and systems level utilization. But right now - I can see more negatives than positive for a competitive SSD market if such a deal were to go ahead.


Everspin quantifies MRAM cache opportunity in AFAs

Editor:- May 12, 2017 - Everspin Technologies spoke about various design ins and design wins for its MRAM in enterprise flash arrays in a conference call re Q1 FY17 results. Sizing the market opportunity the company said that a typical U.2 flash SSD in a storage array can use between 5 to 9 of its devices. And in revenue terms this would mean upto about $1,000 of MRAM revenue in a typical rack designed by its customers.

Editor's comments:- Everspin's first publicly disclosed design win in an AFA system was in 2014 - Skyera - whose buyer - WDC - invested in Everspin in January 2015.


Liqid gets $10 million in Series A funding for PCIe fabric management platform

Editor:- May 10, 2017 - Liqid today announced it has secured $10 million in Series A funding led by Marker Hill Capital. Along with previous seed funding from Kingston Technology, Phison Electronics, ABR Capital Management and DH Capital, the company has now raised $20 million

Datacenter architectures have seen significant advances with the advent of the cloud era but have remained shackled by the motherboard/chassis paradigm, which locks resource allocation to the point of purchase. Through innovations in low-latency fabrics and intelligent software, the Liqid CI Platform addresses this painful and costly limitation by leveraging pools of compute, networking, data storage, and graphics processing residing on the PCIe fabric, managed by Liqid CI software, to deliver transformative results.

Editor's comments:- Among other things the SSD IP amassed by Liqid includes an NVMe successor to the racktop PCIe fabric switch box which PLX was talking about in 2014 - according to a blog about Liqid's composable infrastructure by Robin Harris (StorageMojo).

See also:- SSD software, PCIe SSDs,
VCs in SSDs and storage from 2000 to 2017


BeSang poses question about relative cost of 3D vs 2D nand

Editor:- May 10, 2017 - I thought the SSD industry had a long time ago answered the question - which is more important - the memory or the SSD? - but indirectly it has been asked again in a different way...

In a new post on linkedin Sang-Yun Lee, CEO - BeSang says...

"Micron says 32-layer 3D NAND is about 30% cheaper than planar NAND. However, market price of SSD shows planar NAND-based SSD is cheaper than 3D NAND based-SSD. ...Is 3D nand cheaper than planar nand?"

Editor's comments:- My own comments to this (on linkedin) were much longer than Sang's provocative post. But I'm hoping to learn more from what others say. ...read BeSang's post and comment(s)

Sanity check...

From a competitive viewpoint BeSang has been pitching their own way of getting the memory industry to the next level in capacity at a lower price and greater efficiency than the roadmap which Micron, Samsung etc seem to be pursuing. (And as we've seen in SSD history and semiconductor - the best ideas for innovation come from startups rather than encumbents. Intel itself was once a startup... Whether BeSang's technology will get the investments needed to enable wider adoption still remains to be seen.)

Nevertheless Sang's question is - I think - interesting because underlying it are many difficult realities about the bumpy transition of the memory market to a non incremental (same old predictable spreadsheet) technology ecosystem.

When was it clear that SSD architecture was more important than memory?

Many times as you can see in my classic timeliney blog - sugaring flash for the enterprise. And every couple of years it gets said again in a different way. Here's an example from my 2012 SSD market transitions article.

"Knowing the generic memory type characteristics doesn't give you enough useful info about the SSD's characteristics and limitations any more (2012)."

PS - Unfortuntaely you have to be logged in to linkedin to see the original post and comments (and it has to be working). So if you're deciding if it's worthwhile or if the original post has disapeared by the time you read this in 2027 in the news archive in which case you already know what happened. Here is what I commented on linkedin.

Comparing the price of planar SSDs and 3D SSDs at the same capacity level doesn't lead to a reliable conclusion about the competitiveness of 2D versus 3D nand.

The greater maturity and simplicity of the 2D controller will reduce BOM cost now. But 3D (with its more expensive materials) provides a platform for making a better SSD from the endurance view.

That's something which has user value in many market roles.

When memory production is set against a background of high book to bill then it makes business sense for memory makers who have their own SSD brands to bias the mix towards markets like industrial and enterprise rather than low value markets like consumer.

On the other hand companies like Micron also know they have to keep the consumer markets alive (even if starved) because they are the traditional adopters of future generations of new unproven (reliability) memory.

Feed the fab can't operate if the only customers are industrial or enterprise


That's when my finger slipped on the phone and it got posted. I could have written more. But you get the general drift.

After posting to linkedin and publishing the note above I told Sang-Yun Lee privately that "The clever thing was for you to ask such a naive (sounding) question" - because I anticipated that me and my readers will learn much from responses.

Sang replied and posted this... "Thank you for sharing your comment. I understand that there is no simple answer to this question. Though, I would like to make it very simple without considering other factors. When we just consider 'Cost-per-Bit' or '$/GB,' is 3D NAND cheaper than planar NAND now?

It is an important question to NAND industry because NAND goes to 72-layer already. So, technically it is very mature. In other words, it is not easy for 3D NAND to vertically expand from now on. So, it is right time to ask this question."

See also:- $$D pricing! - connects and disconnects with memory costs


V&G releases Kylin compatible rugged VPX SSDs

Editor:- May 10, 2017 - If rugged 6U SSD cards for military radar and flight recorder applications or fast data acquisition is your thing (as it used to be one of my things about 30 years ago) then you may be interested in 2 new products released by .
  • news image VPX SSD rapid io6U SRIO VPX (pdf)
    8TB MLC / 4TB SLC with 2 lanes of RapidIO host interface, approx 3GB/s sequential R/W, fast secure erase.
  • news image VPX SSD rPCIe6U PCIe VPX (pdf)
    16TB MLC / 8TB SLC with PCIe 3 x 8 host interface, upto 5GB/s read, 4GB/s write, and random R/W IOPS (4K) 200K / 160K respectively.
Other OSes supported include VxWorks, Solaris, Linux etc.

Editor's comments:- for a few years now I've had interesting conversations with Limuel Yap , VP - Global Business Operations - V&G about the various aspects of the custom SSD market and especially the ways in which SSD companies which have a strong focus on the military market create product lines which can be just as interesting as the standard products for publicly known applications that we read so much about on the web.

The ingenuity and problem solving of these companies largely remains unknown for good reasons.

Following up on the new products above Limuel told me - "Developing these products will separate us from other SSD suppliers that only have the standard SSD. This marks down our capabilities to offer all kinds of memory storage most specially for military and data recorders users."


Everspin revs up its enterprise sales channel

Editor:- May 4, 2017 - Everspin today announced the appointment of Annie Flaig as Senior VP of Worldwide Sales.

Prior to Everspin, Flaig was VP of Worldwide Enterprise Storage Solutions Sales at SanDisk. Before joining SanDisk, Flaig was VP for Americas Sales & Channel Marketing at HGST driving revenue and market share growth of HDD and SSDs in the Americas region. And before joining HGST in 2012, Flaig spent nearly 20 years with AMD where she served as the Corporate VP of Worldwide Commercial and Enterprise Sales.

Editor's comments:- this is an interesting evidential marker in the timeline of market readiness and shippability of Everspin's MRAM as it changes phase from being an "emerging" memory technology to one which has definitely emerged.


more readers aren't scared of mice

Editor:- May 2, 2017 - The monthly readership of StorageSearch.com increased by over 100% in April compared to a year ago. Thanks.


SMART files for IPO

Editor:- May 1, 2017 - the parent company of SMART Modular Technologies today announced that it has filed a registration statement on Form S-1 with the U.S. Securities and Exchange Commission ("SEC") relating to a proposed initial public offering of its common stock.

SMART has applied to list its common stock on the NASDAQ Global Market under the ticker symbol "SGH." Barclays and Deutsche Bank Securities are acting as lead book-running managers for the offering; Jefferies and Stifel are acting as book-running managers; and Needham & Company and Roth Capital Partners are acting as co-managers.


Sampling the engineering temperature in Toshiba's never never land

Editor:- April 30, 2017 - Speculation about who might become next owner of Toshiba's SSD and memory business continues... As I told you earlier this year I didn't expect this to be a short story, due to monopoly and other strategic interests.

The latest chapter in this flash novella comes as the number of buccaneers circling around the foundering wreck of Toshiba's corporate galleon and hoping to pluck the semiconductor gold IP fabs from its belly has reduced to 4.

A new blog - Politics Muddy Toshiba Bidding by Junko Yoshida , Chief International Correspondent - EE Times - takes an engineering view of who would be the best or worst of the bidders and asks...

"Are Toshiba's semiconductor engineers angry? Are they discouraged?" ...read the article


Silicon Motion comments on calendar Q1 SSD controller market

Editor:- April 28, 2017 - Silicon Motion today announced results for the quarter ended March 31, 2017. Net sales decreased 12% sequentially to $127 million from $144 million in the previous quarter.

Wallace Kou, President and CEO of Silicon Motion said - "Sales of our client SSD controllers were seasonally weak, our eMMC controllers were seasonally flat and our SSD solutions declined due to NAND flash shortage."


SMART HRS announces 8TB 2.5" SATA SSDs for aerospace

Editor:- April 25, 2017 - SMART High Reliability Solutions today announced new higher capacity models in its HRS-M1HC family of aerospace targeted 2.5" SATA MLC SSDs.
  • 8TB capacity in 9.5 mm height
  • 1, 2 and 4TB drives in 7.0 mm
  • operates at altitudes of up to 100,000 ft
  • designed to withstand shock and vibration up to 1000g
  • R/W speeds are 520GB/s and 500MB/s respectively
"The HRS-M1HC extends our M-Series product line giving our customers in the military and industrial markets a cost effective, high-density option for specialized applications." said Michael Guzzo, GM of SMART HRS. "The M1HC is one of the most comprehensive, security-laden SSDs available from SMART HRS. The fact that we provide it in an MLC based drive gives customers additional flexibility when choosing a solution that best fits their need."

Editor's comments:- This family is also supported by an impressive arsenal of triggerable military erase sequences as order options.


see Swissbit in Vegas

Editor:- April 25, 2017 - When you've got the name of a country in the name of your company does that set expectations about where your natural markets are?

So you might think that Swissbit is stuck in Europe - but the company (which offers COTS and custom memory systems) has been selling SSDs into North America for over 6 years.

If you haven't seen them yet - Swissbit says it will showcase its high-end memory and storage products for industrial and NetCom applications at the EDS Summit in Las Vegas next month.

You might not see many other SSD companies there though. I had a quick glance through the 200+ list of associated companies at this event and could only recognize about 3 others.


now Cinderella industrial systems with "no-CPU" budgets and light wattage footprints can go to the NVMe speed-dating ball

Editor:- April 19, 2017 - A dilemma for designers of embedded systems which require high SSD performance is how can you get the benefits of enterprise class NVMe SSDs for simple applications - which integrate video for example - without at the same time escalating the wattage footprint of the entire attached micro server?

A new paper published today by IP-Maker - Allowing server-class storage in embedded applications (pdf) discusses the problem and how their new FPGA based IP enables any NVMe PCIe SSD to be used in embedded systems to provide sub-microsecond latency using "20x better power efficiency, and 20x lower cost compared to a CPU-based system."

image shows where the FPGA IP fits in the context of an embedded low power system using fast NVMe SSDs

The company says the NVMe host IP - which is now available - can be used in an FPGA connected between the PCIe root port and the cache memory, internal SRAM or external DRAM. It fully controls the NVMe protocol by setting and managing the NVMe commands. No CPU is required. It supports PCIe gen 3 x 8 interface.

Michael Guyard, Marketing Director said that - among other things - applications include:-
  • military recorders
  • portable medical imaging
  • mobile vision products - in robots and drones
...read the article (pdf)

Editor's comments:- Now Cinderella embedded systems with low cost budgets and low wattage footprints can go to the enterprise NVMe performance ball. The new magic - in the form of the FPGA IP released today by IP Maker - has the potential to transform the demographics and class of SSDs seen in future industrial systems.

See also:- optimizing CPUs for use with SSDs, SSD glue chips


low yield at sub 20nm is root of DDR4 shortage says DRAMeXchange

Editor:- April 14, 2017 - Quality problems in DRAMs which have been sampling this year at the new sub 20nm generation from major suppliers is at the heart of the issues discussed in a new - market view blog by DRAMeXchange - which concludes that the contract prrice of 4GB DDR4 DRAM modules will rise 12.5% entering 2Q17.

Avril Wu, research director of DRAMeXchange said - "PC-OEMs that have been negotiating their second-quarter memory contracts initially expected the market supply to expand because Samsung and Micron have begun to produce on the 18nm and the 17nm processes, respectively. However both Samsung and Micron have encountered setbacks related to sampling and yield, so the supply situation remains tight..." ...read the article

See also:- inside SSD pricing, storage market research companies
.
What happened before?
.
SSD ad - click for more info
.

storage search banner

..
SSD news page image - click to  enlarge

Michelangelo found David inside a rock.
Megabyte was looking for a solid state disk.
(see the original 1998 larger image)
..
AccelStor NeoSapphire  all-flash array
1U enterprise flash arrays
InfiniBand or 10GbE iSCSI or 16G FC
NeoSapphire series - from AccelStor

related guides
..
15 years ago in SSD news
In May 2002 AMD announced it was sampling 64Mb Mirrorbit flash with 100k P/E endurance and 20 years retention at 125 degrees centigrade.

The next time we reported on Mirrorbit was in October 2004 when AMD's NOR flash memory was the smallest technology in production at 110nm.

AMD's flash business was spun off as Spansion in 2003 (as a joint venture with Fujitsu).

Spansion went into chapter 11 in March 2009 and emerged again in May 2010.

Then in December 2014 Spansion (and the Mirrorbit flash technology) was acquired by Cypress Semiconductor.

Cypress 's Mirrorbit (in 2017) is a 256Mb is a high reliability, temperature tolerant flash memory with operating temperatures from -40C to +125C.

see also:- charge-trapping MirrorBit NOR flash
..
Memory in China – from Chips to Cloud

Emerging Memory Architectures

Lunch
Editor:- May 5, 2017 - Two interesting things and one very essential thing (above) which I noticed today in the agenda of the forthcoming Memory+ Conference which will take place June 1, 2017 in Shanghai.
..
.
industrial mSATA SSD
industrial grade mSATA SSDs
>2 million write cycles per logical block.
from Cactus Technologies

related guides
..
The industry will learn a lot about the "goodness" of new memory tiering products by stressing them in ways which the original designers never intended.
RAM disk emulations in "flash as RAM" solutions
..
Despite the bewildering range of products in the market - the performance characteristics and limitations of ALL flash SSDs are determined by a small set of of architectural parameters.
understanding flash SSD performance limitations
..
after AFAs? - the next box
Throughout the history of the data storage market we've always expected the capacity of enterprise user memory systems to be much smaller than the capacity of all the other attached storage in the same data processing environment.

after AFAs - click to read rhe articleA new blog on StorageSearch.com - cloud adapted memory systems - asks (among other things) if this will always be true.

Like many of you - I've been thinking a lot about the evolution of memory technologies and data architectures in the past year. I wasn't sure when would be the best time to share my thoughts about this one. But the timing seems right now. ...read the article

..
Targa Series 4 - 2.5 inch SCSI flash disk
.
2.5" removable military SSDs
for airborne apps - GbE / SATA / USB
from Targa Systems
.

related guides

.
.
How does the memory famine impact industrial SSD service values?
Editor:- April 25, 2017 - 17 years ago I wrote a guide - Why I won't publish your press release? - the first of which still applies today - it's not newsworthy.

The context which brought it to mind was a press release earlier this month from Virtium saying it was now offering 2TB (MLC) and 1TB (iMLC) in its 2.5" industrial SATA SSD range.

That kind of capacity isn't news - as we reported the first rugged TB 2.5" SSD sampling on this page more than 8 years ago - in January 2009. (And competitors in the industrial military market have been shipping 8TB 2.5" drives since 2016.)

Is there anything else you can tell me? I asked.

(Like me - readers need to have a genuine reason to be interested.)

I asked about pricing - but the company said there are too many options and didn't want to be drawn on that. I learned the SLC version is rated at 30 DWPD. But that's not a news story either.

Around about this time I was thinking about the impacts of the flash memory famine on different types of companies in the SSD supply chain. (Whether it's a good thing or bad thing - depends which company you are.)

In the 1970s and 80s chip supply volatility due to semiconductor company book to bill ratios thrashing about all all over the place was a constant reality of being in the electronics business. In recent years we had the illusion that the boom to bust days of memory were over - due to better business management and better prediction systems. So when memory prices increased and supply dates went into "unknown" that was a shock to many people.

Continuity of supply and pricing expectations are just as important for industrial equipment makers as avoiding EOL shock and having stable BOMs.

Tactical stories about memory supply and logistics had come up in some recent conversations I had with industrial SSD makers Wilk Elektronik in Poland and Longsys in China.

So I asked this...

Does Virtium (which is headquartered in the USA) get involved in giving price projection promises to its contract system customers?

Scott Phillips, VP of marketing - Virtium told me ""While we can't 'promise' specific pricing into future quarters, we can provide supply-chain and associated pricing visibility to strategic customers up to one to two quarters out, based on NAND flash allocations, associated pricing and cost averaging of materials.

This at least helps customers plan. In other words, even though we are direct with suppliers, it does not completely insulate us from cost increases should suppliers decide to raise prices, but we can somewhat mitigate pricing volatility like what is found in the distribution channel currently."

There you go. Is that more interesting than just another 2.5" TB SSD story? I hope so.

.
.
.
.
Are we there yet?
Editor:- April 7, 2017 - After more than 20 years of writing guides to the SSD and memory systems market I admit in a new blog on StorageSearch.com - Are we there yet? - that when I come to think about it candidly the SSD industry and my publishing output are both still very much "under construction". ...read the article

.
.
.
.
RAM has changed from being tied to a physical component to being a virtualized systems software idea - and the concept of RAM even stretches to a multi-cabinet memory fabric.
what's RAM really? - RAM in an SSD context

.
.
.
.
All the marketing noise coming from the DIMM wars market (flash as RAM and Optane etc) obscures some important underlying strategic and philosophical questions about the future of SSD.
where are we heading with memory intensive systems?

.
.
.
.
I think it's not too strong to say that the enterprise PCIe SSD market (as we once knew it) has exploded and fragmented into many different directions.
what's changed in enterprise PCIe SSD?
.
.
.
.
.
The same memory block may have different ECC codes wrapped around it at different times in its operating life - depending how healthy it looks. And different ECC codes may be used within the same memory chip at the same time.
Adaptive flash care management & DSP IP in SSDs

.
.