click to visit home page
leading the way to the new storage frontier .....
SSD SoCs controllers
SSD controllers ..
pcie  SSDs - click to read article
PCIe SSDs ..
read the article on SSD ASAPs
auto caching SSDs ..
image shows megabyte waving the winners trophy - there are over 200 SSD oems - which ones matter? - click to read article
top SSD companies ..
click to see directory of SAS SSD companies .
SSD interface and glue chips and IP ..
SSD glue chips
Editor:- September 2, 2014 - Seagate today announced it has completed its previously announced acquisition of the assets of LSI's Accelerated Solutions Division and Flash Components Division from Avago Technologies.


LSI is a leading provider of innovative silicon,
systems and software technologies that enable
products which seamlessly bring people, information
and digital content together.
.......... click for more info about LSI
The company's products are trusted by the world's
best known brands to power leading solutions in
the Storage, Networking and Mobility markets.
SandForce SSD processors - click for more info
the award winning silicon driving leading SSDs
from LSI
..... LSI - addresses and links

Corporate Headquarters

LSI Corp
1320 Ridder Park Drive,
San Jose, CA 95131


See also:- LSI - editor mentions on and LSI's SSD page

Editor:- April 30, 2014 - LSI was ranked #3 in the Q1 2014 edition of the Top SSD Companies List which is researched and published by

Who's who in SSD? - LSI

by Zsolt Kerekes, editor - November 18, 2013

For the past year or so I had been wondering if the glory days of LSI's SSD controller technology lay mostly in the past. But I can now understand why it took them so long to complete this new design - which is almost at the integration level of "SSD market in a chip".

Editor's earlier comments:- October 2013 - LSI - ranked #14 in the top SSD companies in Q3 2013 - operates in the PCIe SSD, SSD controller and auto tiering / SSD ASAPs markets.

The company sampled its 1st SSD product in March 2010. It was a PCIe SSD - which is software compatible with SAS - an interface which LSI helped to pioneer. Elements of the IP in the new SSD design came from Seagate and SandForce.

LSI acquired SandForce in January 2012 - and revealed the strategic significance of this shortly afterwards when it revealed that LSI's PCIe SSD product - the WarpDrive - was being oemed by EMC.

LSI's SSDs are small (rather than big) SSD architecture and span a wide range of market applications including:- enterprise SSDs, industrial SSDs and consumer SSDs.

LSI's interim approach to adaptive DSP ECC IP for SSDs - revealed in August 2013 - is to use tactical LDPC - as a try it again strategy for improving data usability.
LSI - selected SSD milestones from 30 Years of SSD Market History

In March 2009 - LSI announced better support for flash SSDs in the latest update to its MegaRAID SAS adapters. LSI calls this new feature SSD Guard - which can anticipate some types of flash SSD failures in RAID 0 configurations and starts rebuilding data on a spare unit.

In December 2009 - LSI announced it is sampling the LSISAS2208 dual-core 6Gb/s SAS RAID-on-Chip IC to OEM customers. It's intended to support the forthcoming PCIe 3.0 specification, currently under development and provide performance levels that meet the needs of next-generation server platforms based on flash SSD storage. The new LSI SAS ROC will deliver performance levels of up to 600,000 IOPS.

In January 2010 - LSI and Seagate announced they have collaborated on designing PCIe SSDs for the enterprise accelerator market - which started sampling in March 2010. At that time - LSI was approximately the 163rd company to enter the SSD market (not counting SSD SoC makers - which would push the score to about 185).

in November, 2010 - Demartek published a sponsored test report (pdf) which compares the performance of SSDs and HDDs in a simulated web server environment when managed by LSI's CacheCade software - which provides SSD ASAP functionality.

Editor's commnents:- The report shows that throughput and access times were improved by at least 3x using a single SSD cache compared to the HDD only situation.

However - it's disappointing that the sizing of the test was not best chosen to draw meaningful conclusions. Because the web content was only 25% larger than the SSD capacity! It would have been more helpful to design a simulated case in which there was at least a 10x or 100x size difference. Because if you can fit all the web content onto an SSD then you don't need the burden of the "cache" software at all - and might get better results by switching it off.

There are case studies going back nearly 10 years which show that SSDs can provide big speedups in web servers. The exact speedup depends on how fast the SSD is. This test report doesn't answer the question - is LSI's CacheCade useful in a realistically scaled environment?

In March 2011 - LSI finally spun off the Engenio systems business - selling it for $480 million to Network Appliance.

In June 2011 - LSI was one of several compatible companies named in FlashSoft's launch of its auto tiering SSD software.

In October 2011 - LSI announced a definitive agreement to acquire SandForce for approximately $322 million. The transaction is expected to close early in the first quarter of 2012. SandForce president and CEO, Michael Raam will become General Manager of LSI's newly formed Flash Components Division.

In January 2012 - LSI announced it has completed the acquisition of SandForce. And LSI also announced that its PCIe SSD product - the WarpDrive - will be oemed by EMC.

In April 2012 - LSI announced details of its new Nytro family of SSD technologies - which integrate and join up several previously standalone elements in its product line in a new unified marketing roadmap.

In June 2012 - LSI demonstrated its SandForce SF-2000 flash controllers working with Toshiba 19nm and Intel 20nm NAND flash memory at Computex 2012 in Taipei, Taiwan.

In June 2013 - LSI demonstrated new features in their SF-2200 controllers aimed at the notebook SSD market. These are:- Opal compliant encryption and DevSleep technology. DEVice SLeeP (pdf) a very low power sleep mode for SATA SSDs.

In November 2013 - LSI launched its 3rd generation SandForce SSD controller family - the SF3700 - which offered native SATA or gen 2 PCIe interfaces - and incorporated adaptive R/W DSP ECC management. described the underlying business concept as "SSD market on a chip" - and described the design as the most ambitious design of a single chip SSD controller in SSD market history.

In December 2013 - LSI announced that it had agreed to be acquired by Avago Technologies Limited in an all-cash transaction valued at $6.6 billion.

In May 2014 - Seagate agreed to acquire LSI's flash business for $450 million.
SSD ad - click for more info
LSI's Nytro - positioning update
Editor:- June 26, 2013 - I had a useful conversation last week with Rob Callaghan who manages outbound marketing in LSI's Accelerated Solutions Division.

When I say "useful" - I mean - useful to me - because I haven't spent much time looking at the details of LSI's family of PCIe SSD products - due to my impression that it won't tell me anything new about the long term architectural direction of this class of product. Having said that LSI's PCIe flash products do provide a relatively pain-free entry point for many users into the world of enterprise SSD storage - and will generate lots of revenue for the company in the short to near term.

Not too bad for a company which I described as "approximately the 163rd company to enter the SSD market" - when LSI announced its entry into SSDs in January 2010.

In that same 2010 news story - I used the headline - "LSI will Compete with Fusion-io" - because it was a useful shortcut to my guess-ahead at what LSI might end up doing. But with the benefit of hindsight - that headline statement isn't strictly true. LSI doesn't compete at all with Fusion-io on a slot by slot level. And at root is the growing fragmentation within the PCIe SSD market itself.

I started to warn about this 2 years ago in an article - don't all PCIe SSDs look pretty much the same? And so the fragmentation within PCIe flash products due to the different roles these products can play within the enterprise - was an obvious place to start my conversation with Rob Callaghan as LSI already has 3 functionally distinct types of products within its Nytro PCIe flash card product line.

Having said that - there are another 3-4 classes of PCIe SSDs for which LSI doesn't have any products at all.

It's a no-brainer to expect that LSI will launch products to fill some of these other gaps too. But it's also clear to me that due to the LSI's preferences and its business comfort zones with controllers, leveraging its RAID and HDD legacy and the need to provide a simple SSD educational interface to channel partners - the company isn't capable or doesn't have the inclination to go after the same kind of high end dark matter user apps which other vendors like Fusion-io and Virident etc can reach - and even if LSI wanted to - they can't get there - because of differences in philosophy in the sizing of their SSD controllers and assumptions about the SSD software operating environment which are conditioned by fundamentally different beliefs about the co-existence of SSD and HDD in the future datacenter.

Anyway - back to my conversation with Rob Callaghan who had been talking to streams of editors etc about LSI's upcoming product announcement (June 24) about more new Nytro stuff. I learned 3 things.
  • the new thing - LSI's Nytro Elastic Cache. One part of flash can be configured as read cache while another part can be configured for write cache and for semi-permanent storage of often used data such as golden images of desktops in VDIs. As a reliability feature - the write cache and permanent data is mirrored on 2 separate regions of flash on the same card.

    I said - that could significantly reduce the amount of flash capacity you need on the card - while offering some of the reliability benefits of mirroring - because you don't need to mirror data which you can easily read again.

    Rob said - you're the first person I've spoken to who got that. (The people I speak to in these interviews know how to schmooze editors.)

    I said - there are so many ways of getting SSD efficiency - each time a company does something like this it means they can make money in the market at a price point where a competitor - using more flash to solve the same problem - is just bleeding their VCs.
  • the thing which LSI's PCIe SSDs don't do yet. I asked if LSI supports low latency automatic mirroring - clustering across multiple PCIe SSD cards - like some of the high-end products from other vendors do.

    Rob said no - they don't. It will come in future LSI products - and Rob hinted that the way they do it may use a different physical mechanism than the method I would have expected. (For me - the norm is PCIe chip level failover supported by PLX).
  • the thing I already thought I knew. The conversation confirmed to me my earlier assumptions about LSI's positioning in the PCIe SSD market - and limitations of their products. (For example LSI's Nytro WarpDrive doesn't have scalability symmetry.)

    Rob said the typical end user of this product would mostly have just a single LSI PCIe SSD card installed in a server. It would be rare for them to have 2. (Editor - So if you want an environment in which every slot can be a PCIe SSD - then you're looking at a different class of product, and different software and a different bunch of vendors.)
Editor's comments:- As I said above - my gut feel is that LSI will do well in the market with revenues from this generation of enterprise PCIe SSDs - in user systems which can be satisfied by fast-enough SSD acceleration in HDD heavy environments. But the company will have to change cars - rather than just switch lanes in the future with both its controller architecture and software base - if it wants to provide efficient solutions for SSD-centric datacenter installations. My guess is - if that does happen - the only viable route would be another significant SSD company acquisition - maybe in 2 years time.

For a company - like LSI - which has already demonstrated that it can successfully leverage a significant SSD IP acquisition - which it did with SandForce - at technical, marketing and business levels - that shouldn't present a fundamental problem. But until it happens you're going to see a lot of confusingly different but parallel courses being taken by vendors in this market - which aren't inter-compatible.

And that will lead to the paradoxical oddity that maybe 2 of the top 3 companies in the PCIe SSD market don't really compete with each other at all.
LSI's new notebook technologies and the petabyte SSD shelf
Editor:- June 4, 2013 - LSI did a really good job leveraging their acquisition of SandForce.

That's the impression I got when I was talking last week to Kent Smith, Sr. Director of Product Marketing who wanted to talk about the new features in their SF-2200 controllers aimed at the notebook SSD market. These are:- Opal compliant encryption and DevSleep technology (a very low power sleep mode for SATA SSDs which I wrote about in an earlier news story lower down this page).

The data I latched onto in LSI's presentation was that according to 3rd party analyst reports LSI's SSD controllers were used in approximately 1/3 of all the flash memory deployed in SSDs in the client and enterprise markets in 2012.

LSI says that even before the start of the current quarter - the SSD world had consumed over 21 million SF controllers.

Now the likely productivity advantages of enterprises using SSD based notebooks were already known even before the SSD notebook market began. And the benefits of having encrypted drives to reduce the cost of exposure to data loss - when a pc goes astray in an airport or is stolen - are exactly the same for SSDs as they were for hard drives - and therefore need no repetition here.

The intriguing thing for me about LSI's new consumer market controllers - as I said to Kent - was the new possibilities that they could open up in really high capacity enterprise arrays used in solid state archives which will eventually replace disk backup and VTLs.

In my roadmap to the petabyte SSD article (March 2010) I observed that one of the missing IPs in the SSD market (at that time) which would be needed to implement petabyte scale physical flash storage in 1U or less - was fast boot SSDs with very low sleep power consumption.

That will enable bulk storage SSD architects to pack an SSD array into the smallest possible physical volume - and leverage a tape library type access architecture at a lower cost of ownership than tape or hard disk. But to be useful in a solid state world - the worst case access time would need to be much faster than the 1 to 2 seconds which was the power on ready time for flash SSDs at the time of writing that article.

LSI's implementation of DevSleep already gives a 400x power reduction in the not needed mode - and Kent told me their power up ready time is about 250 milli-seconds. In my view that's a good enough figure for software architects to start planning around - and it doesn't take much of a stretch to see how that may evolve to get shorter in the next couple of years - if the market puts a premium on this feature.

The target for the bulk storage SSD should be to get 5 to 10 petabytes of virtualized flash into 1U of rack height - at a power consumption level which means that every shelf can have an identical storage density. I had a gut feel it should be do-able - and an attractive market proposition in the 2016 to 2020 timeframe - based on the model I published back in 2010. In some ways it should be easier now - because I didn't anticipate just how good the reliability technologies for MLC and TLC would get - in particular due to the benefits of adaptive R/W.

So - if you have shares in a company which makes tape libraries or disk to disk backup - you've already had many years advance warning that those products will cease to be commercially attractive when the solid state library market gets going.
we're #2 in PCIe SSDs and growing fast - says LSI
Editor:- May 15, 2013 - LSI today announced it shipped over 40,000 PCIe SSDs in the past 12 months - and has been ranked the #2 merchant supplier of enterprise PCIe SSDs in the US, and is the fastest growing vendor in this category according to a recent report by Forward Insights.
new LSI blog on the value of enterprise flash
Editor:- March 14, 2013 - You won't be surprised to see me mentioning a recently published blog by Robert Ober, System and Processor Architect, LSI - about the value of PCIe SSDs in big datacenters - which includes these statements:-
  • "Work/$ is the correct metric (and not crazy expensive $/bit)."
  • "when users say - $8k PCIe card in a $4k server really? - I am always stunned by this"
I'm guessing that the title of Robert's blog - What are the driving forces behind going diskless? Will 100% flash storage make sense in enterprise? - was either inspired by SEO considerations (stuffing the title with value-loaded words for search-engines) or was predetermined before the blog was written.

I prefer this alternative title - suggested by a banner graphic in the blog itself - An $8K PCIe card in an $4k server - huh!?!
NetApp and LSI do that "my software loves your SSD" thing
Editor:- March 6, 2013 - LSI today announced that its Nytro WarpDrive (PCIe SSDs) have been validated for use with NetApp's Flash Accel (SSD ASAP) software.

Editor's comments:- According to NetApp's pdf - "Flash Accel has the ability to keep the cache warm and coherent in the event of disruptive operations and restart caching from the reboot/crash point, rather than restarting from a cold cache."

But it's not as unique in these respects as their document would have you believe - although this suggestion is probably because of when the document was written.

LSI says in its press release that its "advanced off-loaded multiprocessor architecture uses up to 4x less CPU and memory resources than competing solutions".

Now when you see that phrase - off-loaded - in this kind of context - you can be sure that it's a dig at Fusion-io.

The pros and cons - in architectural efficiency and performance - aren't as straightforward as they appear from this subliminal value-loaded phrasing. I discussed these issues a few years ago in an article in FIO's product page here on the mouse site.

The motivational reasons you might choose LSI rather than FIO (or the other way around) probably have less to do with whether you understand or like the way they design SSD controllers (which are evidence rather than motivations of what lies behind their SSD architecture thinking) and instead I think the reasons you might prefer one or other as a strategic supplier would have rather more to do with whether you're comfortable with their different philosophies about the best routes to the future of enterprise storage and, in particular, whether you agree with their speculation of what the destination looks like.

If you're going to be in the same waggon train for 2-3 years - bumping along comfortably together is what's important.
Over-provisioning flash capacity in SSDs - article by LSI
Editor:- January 8, 2013 - Understanding SSD over-provisioning - is the title of a new article published in EDN and written by Kent Smith, Sr. Director of Product Marketing at the SSD controller part of LSI.

Kent's article describes the trade-offs between performance, the percentage of over-provisioned flash capacity and the useful impact of compressible data - which inside SandForce controllers is leveraged to create additional over-provisioning. The interaction between write amplification counter-measures and the benefits of using TRIM commands on performance are also noted. the article

Editor's comments:- there wasn't anything new for me in this article - which covers similar ground to my 2011 article - flash SSD capacity - the iceberg syndrome - which shows how SSD makers leverage capacity to tweak reliability and performance.

But - having said that - I learned about over-provisioning by 10 years of talking about it - with many SSD companies. And some of the things I put in my own article had been gleaned from past conversations with Kent Smith himself when he was at SandForce - as well as various other people in Violin, Texas Memory Systems and Adtron.

I'm guessing that what Kent would have liked to say on OP may have been "trimmed" by a word count limit in his latest EDN article.

So here are some other suggestions for more substantial and ideas packed articles I recommend - which Kent Smith has written in the past for other publications, and which cover SSD controllers from other angles:-
LSI ships 1 million SandForce controllers / month
Editor:- July 31, 2012 - LSI has announced enhanced support for the Ultrabook SSDs market in its SandForce SF-2200/2100 controllers:- enabling lower SSD power consumption, faster boot and support for "virtually all MLC flash product families".

"LSI has shipped well over 10 million SandForce processors and we anticipate our shipment volumes will continue to increase, driven by the exploding demand and lowering price points for NAND flash technology," said Thad Omura, VP of marketing, Flash Components Division, LSI.

Editor's comments:- last week I asked LSI if the improved power saving feature was related in any way to adaptive DSP care. I haven't got an answer yet - so it may be the answer is No.

On the other hand maybe they're waiting for the Flash Memory Summit (in 3 weeks time) before they say more about their adaptive write DSP IP roadmap.
SSD SoCs controllers LSI/SandForce have shipped over 10 million SSD controllers - since 2010 - and they're currently shipping over 1 million per month.
Proximal Data launches AutoCache for PCIe SSDs
Editor:- July 23, 2012 - Proximal Data announced immediate availability of its first product - a software based SSD ASAP - designed to work with PCIe SSDs - and in particular those from LSI and Micron.

AutoCache ($999 for cache sizes less than 500GB) reduces bottlenecks in virtualized servers to increase VM density, efficiency and performance. The company says it can increase VM density upto 3x with absolutely no impact on IT operations.

"LSI and Proximal Data have combined their respective solutions to provide accelerated enterprise storage performance in a virtualized environment," said LSI's director of worldwide channel sales and marketing Brent Blanchard. "Proximal Data's AutoCache, when used with the LSI Nytro WarpDrive PCIe flash card, delivers explosive performance and scalability by lowering data access latencies and resolving the VM density issue that challenges virtualized environments. We are excited to bring this combination to our customers."

Editor's comments:- here are some questions I asked about the new product - and the answers I got from Rich Pappas, Proximal's VP of sales and business development.

Editor:- How long does it take for the algorithms to reach peak efficiency?

Pappas:- It varies by workload, but typically it takes about 15 minutes for the cache to warm to reach peak efficiency.

Editor:- Is the caching only on reads, or is it effective on writes too?

Pappas:- AutoCache will only cache reads, but by virtue of relieving the backend datastore from read traffic, we have actually seen overall write performance improvements as well. This effect is also dependent on the workload.
LSI announces a new technology roadmap for SSD accelerator components
Editor:- April 2, 2012 - LSI today announced details of its new Nytro family of SSD technologies - which integrate and join up several previously standalone elements in its product line in a new unified marketing direction.

In particular LSI is saying that its legacy MegaRAID controllers and software stack can be used as reliable proven launch pads for its SSD ASAP / acceleration software - which is being integrated in new upcoming generations of PCIe SSD cards (now called Nytro WarpDrives) which use LSI/SandForce controllers.

Editor's comments:- in a 2009 storage market forecast I said - "the high end of the RAID controller market is going to disappear" - and I explained why companies in that market - like LSI had to migrate to PCIe SSDs and SSD systems array technology such as SSD ASAPs to satisfy the emerging needs of their oem customers - which in previous decades had been met by RAID adapters and controller chips.

What LSI has done in the past few years is acquire or develop individual pieces of the technology puzzle - and selling their storage systems business Engenio 12 months ago so that they didn't compete with their storage oem customers - was just as important as acquiring SandForce.

I spoke to LSI about the new Nytro technology last week. From the sales point of view they see this as offering affordable SSD acceleration for the masses. So you're going to see low price point fast-enough SSD ASAPs - rather than the fastest.

Other common features in the product line are that the products are bootable, work with legacy SAS software and have minimal load on the server CPU.

LSI will also work to get better integration between the functionality of its SSD controllers and the host cards and caching software. That should lead to better latency and reliability in the future.

difference between LSI and FIO?

What's the single biggest difference you may ask - between LSI and some of the other companies in this part of the PCIe SSD ASAP market? And in particular a company like Fusion-io?

The technical ingredients above are very different - and I could summarize that by saying LSI is at heart an SSD hardware company with most of its IP in chips - whereas FIO is at heart an SSD software company which uses chips as deliverables - but nearly all FIO's IP is in software. That's one way of looking at it - but the clearest difference I see between LSI and FIO is where they are in the philosophy of their thinking re the SSD market adoption model.

All the Nytro marketing orientation materials I saw still talked a lot about how SSDs would fit into an HDD world.

When I questioned that - I got the impression that LSI's corporate marketing hasn't gone much beyond that stage. LSI is still at the "SSDs help HDDs point" whereas FIO and many other SSD makers - and this publication - and many of you too are beyond that and know that the future of all enterprise storage is solid state. The tricky part is navigating safely from here to there.

Finally - Nytro sounds like a good name for an SSD brand - but it's not entirely original.
pcie  SSDs - click to read article A company called Curtis used to sell a 3.5" FC SSD family called Nitro!FC about 10 years ago. See also:- Inanimate Power, Speed and Strength Metaphors in SSD brands
SSD sudden power loss vulnerability guide
Why should you care what happens in an SSD when the power goes down?

This important design feature - which barely rates a mention in most SSD datasheets and press releases - has a strong impact on SSD data integrity and operational reliability.

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.
image shows Megabyte's hot air balloon - click to read the article SSD power down architectures and acharacteristics If you thought endurance was the end of the SSD reliability story - think again. the article

storage search banner

Seagate to acquire LSI's flash business for $450 million
Editor:- May 29, 2014 - Seagate today announced it will acquire the assets of LSI's Accelerated Solutions Division and Flash Components Division from Avago Technologies for $450 million in cash.

The transaction is expected to close in the 3rd quarter of calendar year 2014, subject to regulatory approval.

Editor's comments:- Seagate has for a long time appeared to be a bystander rather than a cheerleader in the SSD market.

By acquiring the SSD controller technology which is at the heart of a significant proportion of SSD designs in the industry - and also the 2nd largest enterprise PCIe SSD maker (by volume) - Seagate will come to be regarded in a different way by SSD customers, partners and competitors.

The cost of the acquisition - seems low compared to many past examples in recent market history.

LSI itself paid $370 million for SandForce in October 2011 - and you might think that in all the years of market and product development since then - and particularly with the controller product line having been upgraded to being an SSD drives and module business too - the combined business should have been worth more.

But it wasn't - and Avago - which has owned LSI's business for a few short breaths of market time - obviously must feel it is getting a good deal.

Possible reasons for the low valuation of LSI's SSD business (in comparison to maybe what it might have fetched a year or so ago are:-
  • LSI was latecomer to the adaptive R/W controller market. So that meant many SSD companies which needed this new type of flash management technology had already made other strategic plans for their next generation flash management roadmaps before LSI began sampling the SF3700 in November 2013.
  • There are strong competitive offerings available now in every segment of the SSD market - and in comparison to the past when a single controller product could effectively be competitive in a wide variety of markets - that's no longer true.
  • Software has become an essential part of enterprise SSD offerings. Not only does it provide essential compatibilities - but it can also add efficiencies, reliability and performance.

    LSI's enterprise software offerings were always weak add-ons in my view - in comparison to competing products - and were tactical sales demonstration tools rather than genuine hard core enterprise platforms.
  • Rackmounts. To succeed in the traditional enterprise SSD market today - enterprise SSD makers need to think like systems companies or become actual systems companies. That's where new efficiencies come together with new architectures.

    But I can't think of a single leading SSD systems vendor today (unlike a few years ago) who would now offer LSI's controllers as the flagship controllers in a new high end array.

    The reasons being that LSI was late to market with adaptive R/W, late to market with big architecture controller design, and doesn't have a scalable software platform.)
In a way none of those valuation theories matters - but in another way they do - because understanding LSI's weaknesses as an SSD company stretched in too many markets takes us neatly onto - what I think could be the real synergy between Seagate and LSI's controller business.

What's Seagate been historically good at?

Making huge numbers of storage drives (hard drives) cheap and affordable.

What has the SandForce controller design been historically good at?

Getting into more design slots for value based enterprise and embedded applications than any other SSD controller design.

My guess is that even if Seagate disregarded any new markets - and focused only on the high volume potential of existing cloud infrastructure customers and big web entities (like Google and Baidu) - who need value based enterprise SSDs - but who are perfectly capable of designing their own software and APIs and firmware tweaks - then Seagate could leverage the LSI SandForce SSD roadmaps for the next several years as a business tool to establish it as one of (several) leaders in the utility SSD segment of the cloud.

LSI on its own wouldn't be able to make such vast quantities of drives at commodity prices.

Seagate on its own didn't have the SSD IP.

But the new Seagate (with SSD IP now in its core) can leverage that IP with its production control skills to cement new price thresholds for value enterprise SSDs.

See also:- hostage to the fortunes of SSD - why big companies can't afford NOT to be in the SSD market...

Afterthoughts (Friday May 30)

PS - Some of the factors dampening the valuation of LSI's flash business seemed so obvious that I didn't mention them in any of the bullet points above. But it appears that they weren't as obvious to everyone - so here they are if you're interested.
  • Competing with your own customers.

    When LSI acquired SandForce - that introduced a new dynamic for the outlook of the combined business - because LSI-SandForce was from that point onwards competing with many of the companies which had previously been customers of SandForce controllers.

    That triggered plans in some of the big customer (now competitor) SSD companies to look for alternative technology roadmaps which would level the playing field in a future when they would be competing - at a drive level - with the controller architecture which they had helped to establish as a de facto industry standard.

    And it also created a more favorable climate for competing SSD controller makers who had lost out before. Such as Marvell.
  • How much are SSD companies worth?

    Generally - SSD companies ain't worth as much as they used to be.

    Nearly all mainstream enterprise SSD companies (who have been shipping products for a few years or more - as opposed to stealthy startups) have seen their valuations dip in the past year - due to investors realizing that the idea of any one company dominating any one big market segment sustainably is an illusion (for now).

    That's because the barriers to entry from new vendors are still low (due to the existence of a very sophisticated SSD IP and parts ecosystem).

    This means that startups using new technologies at low volumes can still deliver SSDs which are significantly cheaper to make than legacy SSD technologies - even when those legacy SSDs are being produced in much higher production volumes. (Legacy in this context - means anything that's more than 12 to 24 months old.)

    I suspect that the industry wide downgrade of the value of SSD companies (which we saw last year with companies like Fusion-io, Violin. Stec, OCZ etc - for various different reasons) will change direction - when we start to see dramatic changes in reported revenues and profitability.

    How will that happen?

    I think it depends as much on creative leadership, vision and discipline in business models as it does on technology.

    Because you can still make money if you sell the right kind of SSD to the right kind of customer for the right kind of purpose. (Especially if they come back soon after to buy more.)

    But you can still lose money by selling an SSD that's better than it needs to be.

    And you can lose money if you over invest sales and marketing resources by selling the right kind of SSD for the right kind of application to a customer who won't need any more of your products for another few years afterwards.
All sounds boringly obvious business stuff doesn't it? That's why I didn't include these add on notes in my original short version of this post.

Another question which it's inevitable to ask - however - is this...

What will happen to the continuing supply of SandForce SSD controllers?

It's hard to imagine that Seagate would - from its own self interest point of view - choose to continue supplying controllers to such a diverse range of SSD companies indefinitely.

Users of these controllers have been on notice that the supply outlook for these essential chips might change ever since LSI acquired SandForce in January 2012.

But unlike LSI - whose main business was selling chips - Seagate's business is selling drives - and Seagate's volume takeup of these chips (for the reasons mentioned above) will be more than enough on its own to justify the NRE costs - without having to sell them elsewhere.

So that may trigger a greater sense of urgency for answers about this factor - from smaller customers who haven't made alternative plans yet.

My guess is that if the regulators look at this transaction - that may be one of the aspects they will focus on - because the contiunuing supply of these controllers to smaller SSD companies in niche markets would a good thing from the viewpoint of competitive choice. (And it wouldn't be too onerous for Seagate to maintain the supply for a time limited period - because Seagate's biggest SSD competitors all have their own controllers anyway.)

Later note added - June 4, 2014

I've been talking to a key contact in LSI's flash business to ask if there was any official corporate statement they could make to clarify the status of this question:- what will happen to the continuing supply of LSI SandForce SSD controllers?

But I wasn't surprised to learn that - despite best efforts - there is nothing more they can say about this for now.

The reason I'm not surprised is because there are significant external entities (regulators, customers, competitors) which could have a bearing on the final decision and if indeed it goes ahead. So it's not just a matter of whatever LSI and Seagate would like to do.
LSI is #2 in PCIe SSDs
Editor:- January 22, 2014 - LSI said today in its quarterly financial report it has shipped over 100,000 PCIe SSDs since it entered this market in April 2012.

Editor's comments:- LSI shipped 60,000 units in the 8 months from May to December 2013 (a figure calculated by subtracting the numbers contained in an earlier announcement.)
Avago wants to acquire LSI
Editor:- December 16, 2013 - LSI today announced that it has agreed to be acquired by Avago Technologies Limited in an all-cash transaction valued at $6.6 billion.

Editor's comments:- I hadn't heard of Avago before. But I had heard of Agilent Technologies - the former name of this company - which was a spinoff from HP. Avago was a semiconductor spinoff from Agilent.

Part of Avago's rationale (they're a semiconductor company with over 1/2 their business in wireless technology) is to get into the enterprise storage market and become a leader in this market "overnight".

A big chunk of the investment - about $1 billion is coming from Silver Lake - a VC company - which a few years ago reintegrated another company in the SSD market - SMART. The rest of the funding is coming from bank loans.
LSI integrates "SSD market on a chip"
Editor:- November 18, 2013 - LSI today launched its 3rd generation SandForce SSD controller family - the SF3700 which - based around a single chip design - spans a wide spectrum of SSD market applications (from consumer to enterprise) - includes native jumper-selectable SATA or gen 2 PCIe interfaces - and incorporates adaptive R/W DSP ECC management.

Editor's comments:- The SF3700 (now sampling) is the most ambitious design of a single chip SSD controller in SSD market history.

Its 14 core design integrates many impressive design and architectural features including:-
  • the ability to efficiently configure as either a small architecture or big architecture SSD controller.

    The SF3700 design can be configured with as little as 3 flash chips in entry level consumer SSDs - or as many as 129 chips when maximally configured in a 9 channel enterprise design which can recover from the complete failure of a memory chip as well as partial failures in other memory chips in the array.
  • dynamically adjusted power islands within the chip - enable a single silicon design to support both the low requirements of deep sleep mode in SATA notebooks as well as the performance requirements of entry level PCIe SSDs.
I recently spoke to Kent Smith at LSI about this new product.

Our conversations about SandForce SSD controllers go back more than 4 years - so we skipped a lot of stuff.

One of the first things I said to Kent - was - I've been nagging you for years and asking - when are you going to do a native PCIe SSD controller?- and for nearly 2 years it's been clear that another big hole in LSI's SSD IP bag has been adaptive R/W - and now you've finally done both at the same time in a single product.

I was also really impressed by the quality of LSI's briefing document on the LSI SandForce SF3700 (pdf) - which explains just about everything you need to know. So I asked Kent - why does he need to waste time talking to editor's like me? - why doesn't LSI just publish the document on the web and let it speak for itself?

I said a lot of publications will simply copy some of your pictures without attribution - and I think readers would find it valuable seeing them too - but I think it would be fairer to the work you've done if I could just make the whole document available - so there was no doubt who had done the hard work of communicating what the design was all about.

He agreed to that - and you can click on the link above to see the original info which I got from LSI.

Some other things I learned from this conversations were:-
  • The SF3700 is a completely new design. - It leverages all the flash related design concepts related to endurance and array level fault management which have been proven in earlier designs and extends them too.

    For example RAISE has been enhanced so that for high-end configurations it can protect against a full memory chip failure as well as multiple block faults - whereas entry level SSDs which need some RAID like features but can't afford an extra memory chip can use fractional RAISE.
  • One of the reference designs which LSI offers for this controller is for an M.2 form factor - which is goiing to be the game changing SSD for the consumer market next year. The card design is the same whether the SSD is being used as a SATA or PCIe SSD. A single jumper sets the configuration at assembly time.

    LSI's 2.5" reference design will also make it easier for oems to produce products for enterprise arrays in the 2.5" PCIe SSD market.
Overall I think the SF3700 is a very ambitious and outstanding SSD controller design - which will elevate LSI's reputation within the SSD industry.

For the past year or so I had been wondering if the glory days of LSI's SSD controller technology lay mostly in the past. But I can now understand why it took them so long to integrate this new design - which is almost at the integration level of "SSD market in a chip".

A design which integrates so many architectural features which are optimized for so many markets wouldn't have been feasible for a small SSD start up.

Below you can see one of the many pictures I spoke about in LSI's paper. If you click on it you'll see the whole thing.
briefing doc on new SF controller
PS - I almost forgot to mention one funny marketing thing I learned.

Kent told me that they used to call the SandForce products "SSD processors" but then found that didn't show up too well in web searches - because people were looking for "SSD controllers" instead. So LSI has changed its parlance and is now calling them "SSD controllers" too.

As I said above - Kent draws some great pictures which illustrate the functional blocks within SSDs. He also writes a lot of SSD blogs too. So I was relieved to hear that he gets paid on an SSD marketer's pay grade rather than that of a writer. That means SSD editors and bloggers don't need to worry that he'll be tempted to come and replace us.
LSI says it pays to get a 2nd opinion from LDPC
Editor:- August 13, 2013 - in a presentation today at at the Flash Memory Summit - the Nibbles and Bits of SSD Data Integrity (pdf) - LSI explained why reserving the use of LDPC to deal mostly with read error retries (and also later in the operating life of flash cells) can be a pragmatic design choice.

And instead of applying different strengths of ECC for fixed physical block sizes - the company says another approach is to have variable sized virtual blocks - which effectively means that better cells carry lower ECC overhead.

See also:- my SSD cell care scheme is better than yours
more articles on
how fast can your SSD run backwards? - 11 key symmetries

You use PCIe SSDs - do you need to look at MCS too? - are these really different markets?

How will hard drives fare in an SSD world? - shrinking into the new gaps created by SSD empires?

Where are we now with SSD software? - (And how did we get into this mess?)

the new SSD efficiency - making the same (functional) SSD - but with a lot less chips

why use SSDs? - user value propositions - the classic article

adaptive R/W flash care management IP (including DSP ECC) for SSDs - what is it?

enterprise SSDs - exploring the limits of the market in your head - is about enterprise SSD futurology.

the Survivor's Guide to Enterprise SSDs - a list of do's and don'ts

the SSD Buyers Guide to SSD guides - includes summary of everything important that's happened in the SSD market in the past year - and has a top level list of SSD articles themed by markets, interfaces and form factors

what do enterprise SSD users want? - and why aren't vendors asking.

The big market impact of SSD dark matter - some of the very biggest direct customer opportunities for SSDs aren't the big name computer and storage oems.

Can you tell me the best way to SSD Street? - I'm like the Old Woman of the SSD Village who talks to everyone that passes through.

comparing the SSD market today to earlier tech disruptions - applying a sense of perspective to what's happening now with SSDs

RAM Cache Ratios in flash SSDs - Knowing whether a flash SSD is skinny, regular or fat tells you a lot. Each SSD type has common modalities.

more SSD articles - sometimes it's hard knowing where to look next...
How big was the thinking in this SSD's design?
Does size really does matter in SSD design?

By that I mean how big was the mental map? - not how many inches wide is the SSD.

The novel and the short story both have their place in literature and the pages look exactly the same. But you know from experience which works best in different situations and why.

When it comes to SSDs - Big versus Small SSD architecture - is something which was in the designer's mind. Even if they didn't think about it that way at the time.
click to read the article - Big versus Small SSD  architectures For designers, integrators, end users and investors alike - understanding what follows from these simple choices predicts a lot of important consequences. the article

STORAGEsearch is published by ACSL