industrial SSDs guide - StorageSearch.com

Indeed an important part of flash SSD history is that it started in industrial applications and only became mainstream and interesting - from a financial investor point of view - when flash SSD designers turned their gaze towards other directions like the consumer and enterprise markets.

A good rule of thumb in the exciting days of disruptive change in the SSD market during 2004 to 2016 was that if you knew the capabilities of leading edge industrial SSD products in any one year the picture was probably very similar the next year too.

There were many good reasons for this. And the predictability and calm, careful approach to new technology adoption in industrial SSDs was widely regarded as a virtue compared to other brasher markets.

But there are many reasons to believe that the industrial SSD market will soon become a new focused melting pot of innovation and architecture.

I describe some of the indicators which brought me to this surprising conclusion in a new blog - not your grandfather's industrial SSD market.

And different ways of looking at reliability is one of the recurring themes which I notice in stories about the industrial SSD market.

Earlier this year I had noticed a statement in one of the customer case studies on the web site of Cactus Technologies which talked about having delivered 200,000 high reliability flash storage cards to a customer "without a reported failure". And from time to time I wondered what did that really mean?

So this week I asked Steve Larrivee, VP Sales & Marketing at Cactus what was the time period behind the story?

Steve said - "The 200,000 cards were delivered over a 2 year period over 5 years ago without one reported failure."

To my way of thinking this is one of the differences in high quality embedded SSD customer care which we don't hear enough about. And the quality of such designs is one of the reasons that customers can still design affordable products which (unlike datadenters) don't rely on live-in service engineers to keep them running.

If you've got more stories like this just ping me and let me know the details.

This article will help you understand why some SSDs which (work perfectly well in one type of application) might fail in others... even when the changes in the operational environment appear to be negligible.

This is my last article on endurance. No more. Ever. I promise. (I have said that before but this time I really mean it.) ...read the article

Interesting to see that the range internal flash memories within this single (superficially fairly simple standard) family includes:- SLC, pSLC (2D and 3D), MLC, and TLC (which is 3D of course).

This shows how sophisticated and nuanced the embedded market has become at analyzing value and selecting the operating parameters for different use cases.

CoreRise's Product Manager said - "Before mass production, there are more than 500 items of the tests in 57 categories to be passed. Moreover, the test standard is very strict. It need not only to conform to the customer requests or nominal standards, but also enough safety redundancy, and guarantee the reliability and consistency of technical performance."

Editor's comments:- The interesting thing in this story is how the customer qualification processes and verification tests for reliable operation in harsh environments for electronics take longer than the original design of the SSD. Thats one of the distinguishing characteristics of the industrial SSD business and sets it apart from consumer and enterprise markets.

From the viewpoint of your company what were the big SSD, storage and memory architecture ideas which emerged and became clearer in 2016?

Here's what some industrial SSD makers said.

"I think definitely it is NVMe" - said a manager at a leading rugged SSD company (who didn't want to be named in this article).

"...page-based-FTL running on DRAM-less controller architectures..." - said Susan Heidrich, Sales & Marketing Manager - Hyperstone

"...Self-encryption, advanced remote-monitoring software... SSDs for Industrial Internet of Things" - said Scott Phillips, VP Marketing, Virtium Solid State Storage and Memory

But some interesting sanity checks came in these replies.

Camellia Chan who is Managing Director at Flexxon said - "Hi Zsolt, Basically we are not the technology leading company. Instead where we are focusing is on continuing to provide legacy product to industrial, medical, automotive customers We support a lot of EOL SSD products worldwide. Most of our customers are facing the problems of discontinued products and we are the ones who support them."

Limuel Yap VP of Global Business Operations - V&G said that many of V&G's efforts in 2016 have been in enhancing previously designed and conceptualized products.

See what others have been saying in what were the big SSD ideas which emerged in 2016?

industrial SSDs

by Zsolt Kerekes, editor - StorageSearch.com

Some of the pioneering companies which shipped those early industrial flash drives are still in the market today. And you could say that the industrial SSD market has remained a constant factor in the fast changing world of the SSD market. But during the past 5 years - the vendor definitions of what makes an "industrial grade SSD" - have changed.

Buyer assumptions which were once safe and universal - can no longer be taken for granted.

Some assumptions have stayed the same.

These are:-

• standard operating temperature range - from -40C to +85C. (Note - some vendors offer an extended range - from -45C upto 90C too.)

• rugged operation:- industrial SSDs must be capable of operating and surviving in a range of vibration conditions typically found in vehicles - and sometimes upto specific MIL standards.

• reliability considerations:- unlike the enterprise server SSD environment (in which individual disks can be backed up, RAIDed or repaired) and unlike the notebook SSD environment (in which flaky firmware can be fixed by downloads and reboots) - reliability and stability constraints on industrial SSDs are much higher. In many embedded applications the assumption is that the SSD lifetime is the same as the equipment life. Industrial SSDs have to be "fit and forget".

• electrical power:- industrial SSDs are available with a vast range of power consumption envelopes (more than 20 to 1) - and in many applications the SSD wattage is a key starting point for selecting a product.

  Industrial SSDs need more complex designs to protect data integrity when there is a sudden power loss - because they are used in a wider range of power supply systems than typical computer systems.

• physical size:- if it doesn't fit in the space then it can't be used. That's one reason why industrial SSDs are available in very small form factors - which I've lumped together in the 1 inch SSD directory.

• OS driver support:- industrial SSDs have to support operating systems which are additional to those used in the computer market - such as real-time OS's

• EOL (end of life) issues. Companies which use industrial SSDs for use in equipment and embedded systems need to be confident that they can continue obtaining these products during the lifetime of their programs - which are typically 3 to 5 - but can also be more than 7 years (see article below). That's dramatically different to the consumer and enterprise SSD markets where product lifetimes are typically in the range 6 to 24 months.

  The reason is that industrial SSD customers want to minimize the amortized costs of requalifying new products and doing redesigns in what are typically low volume markets. Consequently industrial SSD companies have to manage their technology roadmaps and customer service interfaces in a different way to computer market SSD makers. This is similar to the requirements in the military market.

Some important assumptions in industrial SSD product families have changed since this market began.

• use of MLC flash memory. Not that long ago - it would have been unthinkable for industrial SSDs to be offered with anything other than SLC flash memory. The industrial SSD market was slower in moving to an MLC permissive market than the enterprise SSD market because of the need to qualify other associated components - not just the flash chips - over the wider temperature range.

• Integrating SMART-like controller metrics back into the equipment design loop. Most industrial SSD makers offer tools to help equipment designers understand - the on-board SMART-like data collection features in their SSD controllers - and how use this info to model system life.

• Performance. Until recently - the fastest industrial SSDs used to lag 3 years or so behind similar form factor SSDs in the computer market. Not anymore. They are now available with speeds which are similar.

• Encryption and security. The range of security options and the associated economics in the industrial SSD market are as diverse as the markets in which they are used. For example the storage zones which hold the programs and other critical data can made unreadable outside of defined conditions to prevent competitors reverse analyzing the design.

• Industrial SSDs used to be replacements for hard drives. But in new designs we're seeing embedded SSDs replacing functions which were once done by vanilla flash memory. Why's that? It's because when memory capacity gets above a certain size - then the only reliable way to manage it is with SSD-like controller technology. So designs which once used bunches of flash chips are morphing into slots for low end SSDs - simply as a pragmatic way of using flash without equipment designers needing to peek inside unreliable flash chips.

Industrial SSDs are 1 of the 5 main SSD user value proposition groups described in StorageSearch.com's SSD market penetration model. But this isn't a market which shouts loudly about the latest technology fashions. And it's fair to say that because of the conservative and largely hidden nature of the market - it rarely grabs the storage news headlines. But the industrial SSD market has great potential. One SSD analyst - Web-Feet Research - said recently they thought that "the embedded flash drive market would approach $17 billion annual revenue in 2015."

To find related companies, products and related articles - see the list on this page.

• the SSD reliability papers
• a simple list of military SSD companies
• EOL related issues for industrial SSD BOMs
• principles of bad block management in flash SSDs
• where are we heading with memory intensive systems?

He said he'd been having difficulties. Anyway I suggested he look at the industrial SSD directory (this page) and contact all the companies to see who did have such a product and after a lot of work he has found some possible suppliers - but the success rate comes out as about 20% - or 1 in 5 industrial SSD oems actually having a useful MLC product in this category. So this is still a part of the SSD market where SLC reigns supreme - regardless of what you hear about in the server and consumer markets. (And the other SLC SSD bastion is military.)

I said - I think that one of the problems in finding cheaper MLC SSDs for industrial temperature apps is that the few companies who really have the IP to be able to design such a product from the controller stance - mostly pursue bigger markets - like consumer. And even if the consumer guys do have the controller technology for extended temperature MLC - they don't cope well with other factors - like surviving SSD sudden power loss. But it's good to know that some companies can do it - even if it's less than you would expect.

A few hours after posting the above - I got this helfpful comment from another reader (on a different continent)...

Peter Kindl, Managing Director at Solid State Storage Consulting in Germany emailed me to say -

Hi Zsolt, I'd like to comment on your today's news "SLC still rules in industrial SSDs" and your article about industrial grade SSDs.

The most critical criteria for industrial clients is neither of technological nor technical nature, but a process management topic: industrial clients expect a true fixed BOM management by the supplier, in order to get exactly the same product, that they had qualified for their system.

Fixed BOM has become a sales & marketing argument and is being misused. Many suppliers allow "minor" changes, i. e. F/W updates or even dye changes under their "fixed BOM" definition, which can cause severe integration issues at the client's site.

Digging deeper into that subject, fixed BOM suddenly turns out to be marketed the same as so called controlled BOM for certain suppliers, whose core business is still the commercial or client SSD market.

In addition, it's my experience, having been in that market since 2005, that the typical life time expectation for SSD product is closer to 7+ years - at least with most German industrial clients (telecom, IPC and industrial automation). And in many cases, you start the evaluation with a current product generation, that will be almost EOL by the time the client has finished his qualification.

Therefore, sharing the product roadmap and migration plan becomes an important issue, in order to minimize re-qualification cost. This is also becoming a major challenge as most Asian suppliers are hesitant to even talk about current technological matters like NAND flash supplier and lithography being used in the product as well as the NAND flash controller manufacturer.

The top tier industrial clients in Central Europe have a certain expectation by history of the past 5-8 years and new comers don't seem to understand the market requirements when they enter that market. At this year's Embedded World in Nuremberg, Germany it became very obvious, that many new and "old" suppliers (Kingston!?) want to get a decent piece of that tasty cake.

Best regards - and keep up the good work at storagesearch.com