click to visit home page
leading the way to the new storage frontier .....
image shows software factory - click to see storage software directory
SSD software ....
read the article on SSD ASAPs
auto tiering SSDs ....
enterprise SSD segmentation
enterprise segments ..
image shows megabyte waving the winners trophy - there are over 200 SSD oems - which ones matter? - click to read article
top SSD companies ..
SSD SoCs controllers
SSD controller chips ..

FlashSoft - circa 2011

Founded in 2009 and based in Mountain View, CA - FlashSoft - will be the premier software provider enabling the intelligent use of flash storage in the enterprise.

FlashSoft is introducing ground breaking solutions that enable efficient use of flash SSDs and flash PCIe cards in the Windows, Linux and Virtual marketplaces.
.... FlashSoft logo - click for profile

see also:- FlashSoft - mentions on, SSD history,
acquired SSD companies, 90% of enterprise SSD companies won't survive

Editor:- June 2016 - FlashSoft was one of those rare SSD companies whose technology and brand continued to be developed many years after having been acquired. (This is being written 4 years after FlashSoft was acquired in February 2012.)

And I believe the insights and contacts in the enterprise SSD market - which came from having direct contacts with users of its enterprise software at a critical period of change helped its purchaser - SanDisk - to evaluate important user trends in the market and better adapt towards an enterprise friendlier way of thinking.

How much of that - if any - will feed through into adoption by SanDisk's new owner - Western Digital - remains to be seen.

From a legacy point of view it would be much simpler for WDC to seek hybrid and cloud-like systems level software by another acquisition - such as Tegile - rather than integrate all the various software jigsaw puzzle parts which come with the component level IP sets of SanDisk (which include Fusion-io and Schooner Information Technology for example).

A critical test is how WDC plans to position itself within the emerging SSD and SCM DIMM wars market.

This is market it could have chosen to stand aside from - before it acquired SanDisk. But the big memory market is central to the future profitability of flash (and WDC now has skin in that game) and the big memory market also intersects and competes with product lines (such as PCIe SSDs and 1st generation memory channel SSDs) in which SanDisk has had iconic products in the past. (I expect some of those legacy product lines to be end of lifed.)

SSD aware software - rather than interface and controller chip design - has become the new heat pump in the entropy of data systems. The outline challenges and complexities of the big memory market are still not widely understood. That's why I made this difficult theme the subject of my recent home page blog - where are we heading with memory intensive systems and software?
SSD ad - click for more info

Who's who in SSD? - FlashSoft

by Zsolt Kerekes, editor, - December 2011

FlashSoft is 1 of 30 companies which has launched SSD ASAPs, and was 1 of the first 2 software companies to ever be listed in's long running Top SSD Companies List - based on reader metrics in Q3 2011.

As you can see from the company's current description of itself (above) it's clear today what the company is doing / aspiring towards.

But when Ted Sanford, CEO of FlashSoft first contacted me (while in stealth mode and more than a year before initial product launch) from his cryptic gmail address - to discuss the market for his new auto tiering / SSD ASAP software company - my immediate (mistaken) reaction on seeing his email was - "Oh is this going to be another company looking at hybrid notebook SSDs?"

But I perked up when he went on to explain that - no it wasn't about notebooks - their product was aimed squarely at the server acceleration market.

From what I learned then - it seemed to me that they might be able to make a significant difference to the dynamics of how SSDs were being sold and used. If their product worked and if their marketing was worked too.

I knew that lots of other companies were working on this problem too and competition would be tough. Over 30 companies have since announced SSD ASAP related products - and I think this type of technology will eventually be marketed by hundreds of vendors.

In the 2 years since the earliest SSD ASAP products came to market - I've upgraded my view about where they fit into the storage landscape.

I now think these products will have a permanent place in the storage market – instead of merely being a stop-gap product (while HDD arrays are still in use in the enterprise).

FlashSoft was the first company I spoke to which had also independently reached the same conclusion.

It's unusual to suggest that a startup might have a vision of where its technology would fit into the SSD market in 5 years' time - but from the strategic point of view - I think that unless SSD companies are clear about that kind of thinking right from the start - they will often be reactive and off target and waste resources in directions which prove to be futile.

I used the term "its technology" to differentiate between technology and the legal entity which is the company. The stats from storage history indicate that successful new storage software startups get acquired - while most of the rest go bust.

How should you view SSD ASAPs?

The mental model I'm offering readers today is you should think about ASAP IP in the same way you think about SSD controllers. There are useful analogies to be drawn from that market.

Although there will be a small proportion of SSD oems who develop or acquire their own proprietary ASAP technology - I think most of the hundreds of vendors who want to be in the SSD market (because it sells more SSDs) won't find it economic or sustainable to develop their own technologies in-house. Instead they will source solutions from a merchant market for ASAP software, IP and chips – for the same reasons that oems buy 3rd party SSD controllers. (Speed to market, lower risk, lower investment cost, and better ROI while volumes are still low.)

Within the SSD ASAP market I think FlashSoft fits into these classifications:-
  • small architecture rather than big architecture.

    I'd better clarify what I mean by "big architectur" in the ASAP context.

    In the case of an SSD controller - the dividing line is the number of flash chips around which the design is optimized. This affects the entry level product size and also the scaled up economics (when reliability is factored in) and theoretical limits to performance (when using the same number of chips in a customer's rack).

    Here in the context of ASAPs I think the dividing line for big architecture - is one that starts with multiple levels of SSD storage, connected to tens or hundreds of servers - and petabytes of SSD storage. There are no real examples of this product classification which exist yet. But there will be.
The current version of FlashSoft's product supports a terabyte of flash cache using a skinny RAM cache of only 150MB in the server. It uses 3% to 5% of the host CPU cycles. Ted Sanford told me their next Linux release will support upto 256TB flash.

FlashSoft has got the usual stuff on their website such as benchmarks, tech overviews etc.

When I asked yesterday - is there anything in particular you want me and my readers to know? - the 1st thing on Ted's list was - to please mention their October 2011 - release of version 2 software for PCIe SSDs.

He said - "I know you've heard this info, so I won't repeat it, but we think it's a big milestone for the company, especially delivering a 2.0 version before any other software company has brought out their 1.0 version."

Last week - at the Server Design Summit - Ted gave a presentation - and he told me - "I think it reflects our vision for ASAP soltions.

"Today SSD vendors are generally thinking in terms of a single server architecture. We took the lead in developing caching that would run in a single server, but as you can see from this presentation, we built our first products as the "base" of a "technology pyramid" that would lead to clustered environments, integration between server-tier flash and storage arrays and virtualization evironments."

He sent me his presentation along with his speaker notes and I think it gives the clearest picture of what FlashSoft is aiming at.

You can judge for yourself by looking at this overview - Why your next server will have a solid-state cache (pdf)

For related info take a look at the links above and - FlashSoft - editor mentions on

I currently talk to more than 300 makers of SSDs and another 100 or so companies which are closely enmeshed around the SSD ecosphere - which are all profiled here on the mouse site.

I learn about new SSD companies every day, including many in stealth mode. If you're interested in the growing big picture of the SSD market canvass - StorageSearch will help you along the way. Many SSD company CEOs read our site too - and say they value our thought leading SSD content - even when we say something that's not always comfortable to hear. I hope you'll find it it useful too.

SanDisk (enterprise) was a prominent advertiser here on in the years upto its acquisition by WDC. The product line they promoted here was the FlashSoft software IP. You can see one of several ads from the 2014/2015 campaigns here below. In later years the whole SSD market became much more receptive to the idea of big memory and replacing DRAM with flash and other alt nvms. See:- where are we heading with memory intensive systems? for more about that strategic trend.
FlashSoft mentions in SSD market history

In June 2011 - FlashSoft announced it has secured $3 million Series A funding and has launched its first product - software which enables enterprise flash to be used as a cost-effective, server-tier computing resource (ASAP functionality in software) which is available for free evaluation through a 30-day "Try Before You Buy" program.

In October 2011 - FlashSoft became the first ISV to enter the top 20 SSD companies list.

In February 2012 - SanDisk announced it has acquired FlashSoft - one of the leading independent software vendors in the SSD ASAPs market.

  • editor's comments:- October 2011 - FlashSoft was one of the first two software companies to ever be listed in's long running Top SSD Companies List.

    FlashSoft was ranked #14 - based on SSD search metrics in the 3rd quarter of 2011. That was slightly ahead of IO Turbine - the only other software company to have achieved this accolade. IO Turbine would have entered the top 20 at #16 in its own right - if it had not been acquired before the list was published - by Fusion-io.

    FlashSoft designs a software product which provide auto-tiering / SSD ASAP functionality when used with almost any industry standard vanilla flash SSDs. FlashSoft's tagline for this - which I like - is - "Software that unleashes the power of enterprise flash". Another way they describe it is as a "tier minus one" software solution. I'm not so keen on that description - but I suppose it will stick. See why I tire of "Tier Zero Storage"

    FlashSoft was founded by a team with decades of experience at companies including VERITAS, Oracle and Symantec. Having good connections - FlashSoft has been quick to get off the ground and establish technical liaisons with a bunch of leading SSD and server infrastructure partners.

    The exciting prospect for vanilla SSD oems is that FlashSoft's new product - if it works effectively - will make it much easier for them to sell SSDs to new customers who haven't been economically attractive to SSD vendors in the past for the reasons I described in my article - analyzing the emerging market for SSD ASAPs.

    If it's such a good idea then why aren't other companies doing the same thing?

    Well Microsoft did try. They decided not to pursue it - and sold their technology to Adaptec who used it as the core of their MaxIQ SSD Cache.

    Other companies in the software only ASAP space - include IO Turbine, NVELO and NEVEX Virtual Technologies.

    I spoke to FlashSoft's CEO Ted Sanford (June 29, 2011) about their business plans and technology.

    FlashSoft's value proposition is the classic Auto-tuning SSD Accelerated Pools of storage (ASAPs) case:- which aims to provides an economic and simple to install tool to enable servers to get the benefits of SSD acceleration without the costs of either:- migrating all the data to SSD or the high cost of human engineered tuning (which can go out of date). The company says it has achieved worthwhile 4x to 5x speedups in SQL apps which are documented on its web site.

    FlashSoft doesn't yet ship software for clustered servers - but that's in the pipeline.

    How much does it cost?

    At launch time a single server license for a fast SAS SSD is about $1,500 - while a PCIe SSD license costs a shade over $2k. (Have they tested their product with a Fusion-io card I hear you ask? - Yes - among many others.)

    I think the product will be attractive for SSD oems - who can easily tweak the code to make it work better with their own products and open new markets which would remain closed without the simplicity of the ASAP concept.

    Summary - I started talking to the company a year ago when they were still in stealth mode. The direction they've taken and the execution of their launch plan demonstrates a lot of business maturity - which is rare in software start ups. Many companies in the past have tried and failed to convincingly break though the technical and market acceptance barriers in the auto tiering SSD market. This product won't fix all problems - but if their execution remains on track - I'd say that FlashSoft has a good chance of doing in SSD software - what SandForce did for SSD controllers. It will make it easier and cheaper for a new wave of server oems and storage integrators to expand the reach of SSDs in the enterprise.

    FlashSoft will also work with RAM SSDs

    Most of you won't be interested in this bit - which is why I left it till last. But regular readers may know from my SSD branding series that I'm interested in the subject of names and brands in the SSD market - not just in the technologies and markets. And another thing I said in my meeting with FlashSoft that I thought they had chosen a good name for the company - because it describes what they do.

    But I also pointed out that some people may think from the name that their solution is only applicable to flash SSDs. (Just as the the wrong signal gets sent when readers see "RamSan " etc.)

    FlashSoft agreed that their software is not intrinsically restricted to working with flash SSDs - because they already work with latencies in the 10 to 20 microseconds range with flash (even though with flash - the write times are to cache rather than directly through to flash media). Bottom line there's no technical reason why their hot spot tuning software shouldn't work equally well (or better) with RAM SSDs too.

    Once the company has got further along with clustering support - and supporting more operating systems this could make it easier for RAM SSD vendors to get into some applications too. Having said that - the economics depends on the topology of the SAN. If you've got 20+ servers sharing data on the SAN it may still be cheaper to put a single specialist SAN SSD appliance in the path to that data - rather than buy 20 sets of local SSDs and 20 sets of licenses.

    In the long term I think that some customers will do both. Use (less) locally accelerated servers - and still need faster RAM SSD layers above that. And even when all storage becomes solid state - something which FlashSoft and I agree about is that there will still be a vastly different speeds of installed solid state storage in the enterprise. So the need to manage performance between the apps level server and the outer layers of bulk storage will still be there in 5 to 10 years time. A nice kind of problem to worry about - if you're a start up.

Other articles on - which you may be interested in

storage search banner

after AFAs - what's next?
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 thought provoking blog on - cloud adapted memory systems - asks (among other things) if this will always be true. the article
SSD ad - click for more info
One cache to bind them
Editor:- April 2, 2015 - SanDisk - which in the past 4 years has acquired 4 companies listed in the Top SSD Companies List (published by - recently announced the availability of acceleration bundles which integrate 3 of the most recent. Among other things these include:- "We have built our enterprise flash business to offer the broadest range of storage acceleration solutions to address customers' changing needs," said Ravi Swaminathan, VP and GM, Systems and Software Solutions, SanDisk. "The new hardware-software bundles offer the best of both our hardware and software portfolios in a single solution."

Editor's comments:- The new FlashSoft bundles will appeal to some types of enterprise users who like this style of roadmap continuity - where the personality of their SSD (software and controllers) come from the same source.
"Across the whole enterprise - a single petabyte of SSD with new software could replace 10 to 50 petabytes of raw legacy HDD storage and still enable all the apps to run much faster..."
meet Ken - and the enterprise SSD software event horizon
"Many enterprise users - who wouldn't dream of approaching SanDisk to use its raw SSDs - seem more than willing to use SanDisk's enterprise SSD software (FlashSoft). Can SanDisk leverage this to transform itself into an enterprise SSD heavyweight?"
Editor's comments:- SSD news - June 19, 2013
image shows  Spellabyte's software factory
Where are we now with that SSD software?
is SanDisk really nurturing true enterprise SSD DNA?
Editor:- August 15, 2012 - Do you remember FlashSoft?

Many of you still do. It was one of the top enterprise SSD software companies before it got acquired 6 months ago by SanDisk.

One of the tips in the Survivor's Guide to Enterprise SSDs - is that when it comes to SSDs - rules are made to be broken.

And earlier this week I learned this can apply to my own gut feel rules of thumb too. The unwritten rule being that semiconductor companies generally make a mess of enterprise software and are not so hot at understanding the enterprise SSD market either.

Frankly I had expected that FlashSoft would disappear into SanDisk - and would get smothered by a marketing organization which had many times before demonstrated its lack of awareness of the fundamentals of good enterprise SSD marketing. And that was the tone of my parting message to the founders along with a few words of congratulations as they disappeared into the new SNDK afterlife. I never expected to hear from them again.

So the first thing I asked Rich Petersen - (former VP of Marketing at FlashSoft and now Director, Marketing Management at SanDisk) a few days ago was - how are they doing as part of a chip company? What are they doing with the FlashSoft brand? How do they plan to develop the enterprise SSD business? etc.

One of the things that Rich had wanted to talk about was the release of new support in their caching software for VMware vSphere. We spent a lot of time talking about that too - and had a big discussion about the role of SSD software - not only as a business tool - but in effect as a new way of virtualizing and looking at enterprise SSDs and how they can fit into architecture models. (My view is that a powerful SSD suite - if it becomes widely used - can be as significant to the SSD market - as a new interface or form factor.)

We covered enough ground to write several long articles. I'm not going to do that today - because I'm supposed to be on vacation and sitting out in the garden by my pool.

So you should regard this as the really really short version - and a placeholder for much more detail which I will return to later.

FlashSoft - or the enterprise SSD software part of SanDisk (or whatever else you may want to call it) is today operating in a business mode which is like what you would expect from a best of breed enterprise SSD systems company. They talk to end users like they've always done. They learn to change important aspects of how the products work and are sold because of feedback from end users - and not because they've read that something is a good idea in a market analyst's report.

There are some surprising consequences of this at the technical and business level.

Chief among those surprises for me is that FlashSoft says it will still support other brands of SSDs. Rich explained this was just a pragmatic business decision. Big users told them they like FlashSoft - but they already use or might want to use non-SanDisk SSDs. These users are only going to standardize on one SSD software platform. They don't want to learn 2 different ways of doing the same thing.

On the other hand an advantage of having access to an enterprise SSD maker is that if a big user needs some expensive hardware on which to evaluate the benefits of their software - then it's easier on the marketing budget to get some SanDisk SSDs to do this.

FlashSoft's visibility into what enterprise end users really do - and the suprising preferences they have - which are driven by customer business optimizations rather than simplistic technical extrapolations - also means that - like rackmount SSD companies - FlashSoft learns valuable market lessons which can be reapplied to optimize designs in future SanDisk enterprise silicon.
SSD ad - click for more info
SanDisk acquires FlashSoft - but...
Editor:- February 15, 2012 - SanDisk today announced it has acquired FlashSoft - one of the leading independent software vendors in the SSD ASAPs market.

SanDisk says it intends to sell FlashSoft's products as standalone software, as well as offer these software products in combination with SanDisk's growing portfolio of SAS, PCIe and SATA enterprise solutions.

"The acquisition of FlashSoft represents an important step in SanDisk's strategy of delivering complete SSD and software solutions to enterprise storage customers," said Sanjay Mehrotra, president and CEO of SanDisk. "FlashSoft's software products complement our growing family of SAS, PCIe and SATA Enterprise SSDs, and we are pleased to welcome FlashSoft to the SanDisk family."

Editor's comments:- I'm not surprised that someone has acquired FlashSoft - because they were an obvious target sitting so high in the Top SSD companies list.

This means that SanDisk now joins an impressive roster of enterprise SSD makers who have acquired auto acceleration / virtualization software companies in the past year. The only stumbling block is that acquiring enterprise SSD assets isn't the same as being able to do anything useful with them afterwards from the business point of view. Especially when they're software companies.

I still remain unconvinced that SanDisk has achieved as much as it should have done from its earlier acquisition of enterprise SSD controller maker Pliant.

(Although Pliant made SSDs - they had virtually no market share - so their main value to SDK was as a sounrce of enterprise controller IP.)
SSDs over 163  current & past oems profiled
SSD news
I don't think SanDisk understands the enterprise SSD market in the same way as the other companies with which it competes. It's not the same as marketing consumer products.
FlashSoft unleashes the power of enterprise flash
Editor:- June 28, 2011 - FlashSoft today announced it has secured $3 million Series A funding and has launched its first product - software which enables enterprise flash to be used as a cost-effective, server-tier computing resource (ASAP functionality in software) which is available for free evaluation through a 30-day "Try Before You Buy" program.

FlashSoft says that despite the performance advantages of flash SSD, 2 barriers have inhibited its adoption in the enterprise.
  • First, when used as primary data storage, flash memory cannot easily integrate with and leverage the benefits of existing storage systems infrastructure.
  • Secondly, storing all of an application's data on server-attached flash memory remains expensive.
FlashSoft's new all-software product overcomes both of these objections with what they call a "tier minus one" solution for flash virtualization. Enterprise IT can now provide databases, applications and virtual machine environments with the performance benefit of having the entire data set on flash, with only a fraction of the data actually stored in flash.

This innovation makes enterprise flash a cost-effective performance solution that works seamlessly with existing storage infrastructure. In fact, FlashSoft actually reduces the IO burden on storage, producing even greater cost savings. FlashSoft's technology is designed to deliver flash-grade performance within a standalone server, across server clusters, and throughout the data center. Early Customer Successes

One early user - Zenprise said "By using FlashSoft, we aren't buying new server hardware or licensing additional server software. We're simply making our existing servers and software run at their full potential." And they were equally equally impressed by FlashSoft's reliability when they set up stress tests (read case study).

In conjunction with its funding announcement, FlashSoft announced that it is collaborating with industry leaders including VMware, Microsoft, SanDisk, Virident Systems, LSI, OCZ, and server maker AMAX. These relationships will help FlashSoft integrate its software more closely with complementary hardware and software products, and provide customers with the best solutions for their specific requirements.

Editor's comments:- FlashSoft says its software (which runs on Windows Server - Linux is in beta) works with any flash SSDs upto 1TB, and takes approx 5% CPU utilization and 100MB of core RAM. I asked Ted Sanford, founder/CEO some questions.

How many physical SSDs does the software support?

The number of SSDs is not limited, as long as they can be represented as a single logical volume, eg. through a RAID.

Is the 1TB limit shown on your site the limit for the setof SSDs or just for each drive?

The 1TB limit is the current logical limit for the SSD used for caching. The data set is typically 5x greater (or more) than the cache. The size restriction is an artifact of early development, and in a near-future release, there will be no restriction on the size of the SSD employed.

In the case of sudden power loss what are the steps taken to protect the state of the cached data and update the external storage?

FlashSoft employs a method called multi-level metadata management, which stores some cache metadata in RAM, but most of it on the SSD itself (and employs a balanced tree design for optimal efficiency). There are two benefits to this design: first, it minimizes utilization of server memory. Only the hottest metadata runs in server memory. The rest is cached in SSD. Also, the application regularly creates snapshots of the metadata on the SSD, so that in the event of a server crash, the cache metadata can be re-created from the snapshots + most recent metadata almost immediately.
read the article on SSD ASAPs Typical recovery is less than a second. (Keep in mind, our team's background is at Veritas, Oracle, Symantec, etc. so data recovery is a top priority for the product
1.0" SSDs 1.8" SSDs 2.5" SSDs 3.5" SSDs rackmount SSDs PCIe SSDs SATA SSDs
SSDs all flash SSDs hybrid drives flash memory RAM SSDs SAS SSDs Fibre-Channel SSDs is published by ACSL