A second-source for HyperCloud ?

Could IDTI be licensing or second-sourcing RDIMM-compatible HyperCloud ?

I would like to thank one of the readers for suggesting that IDT maybe entering the LRDIMM/HyperCloud space. While IDT intention to deliver LRDIMM in late 2012 was known, the comment sparked an examination of what options IDTI would have if it were actively to use HyperCloud for it’s product and whether it would justify IDTI decision to not offer any LRDIMM product for the Romley launch.

Netlist will eventually face the problem of second-source for HyperCloud.

Currently they have the capability of supplying most of the demand for HyperCloud – both at the 16GB and 32GB levels – they have capacity for millions of memory modules at their facility.

However, they will eventually be required to establish a second-source for HyperCloud by the OEMs.

Background on IDTI

IDTI and Inphi are the top 2 buffer chipset makers.

Texas Instruments is the third – but it has been not interested in the LRDIMM space (possibly because of settlement in Netlist vs. Texas Instruments) – as examined here:

https://ddr3memory.wordpress.com/2012/05/24/lrdimm-buffer-chipset-makers/
LRDIMM buffer chipset makers
May 24, 2012

IDTI behavior regarding LRDIMMs has been peculiar to say the least.

IDTI was in their words the first producer of a JEDEC-compliant LRDIMM buffer chipset. They also scoffed at Inphi claims to have created the LRDIMM standard at JEDEC.

http://ir.idt.com/eventdetail.cfm?eventid=95812
May 9, 2011 1:30 PM PT
IDT Fourth Quarter and Fiscal Year End 2011 Financial Results

DISCLAIMER: please refer to the original conference call or transcript – only use the following as guidance to find the relevant section

at the 26:40 minute mark ..

Blain Curtis of Barclays Capital:

Hey guys. Thanks for taking my question. If you can just talk about some of the strength (?) being in servers, where you saw last quarter and then for the guidance – are you (unintelligible) with LRDIMM and how are you looking at that rollout going forward ?

Answer:

We are seeing very good strength in servers. Intel reported very good strength in servers, so unit demand is going up and our content is going up as I covered in the prepared remarks.

Right now the action is in DDR3.

LRDIMM is really a 2012 story.

We have sampled the world’s highest performance and in fact the world’s only JEDEC compliant LRDIMM and that is sampling today.

However .. uh .. if I was you I would focus the attention on DDR3 because that is really the .. uh .. the technology that is rolling out today.

I mentioned in the prepared remarks that we’ve got increasing market share in DDR3 .. uh .. a year ago on the same (conference) call I remember we were all sitting here and there were a lot of questions about whether or not IDT was going to lose share in DDR3.

Uh .. that clearly has not been the case – we have increased our market share to well above 50% in DDR3 .. uh .. now there are similar questions about whether or not we are gonna lose share in LRDIMM and it’s kinda like deja vu all over again.

at the 28:10 minute mark ..

In fact, it’s a little bit amusing to me that our primary competitor in .. in LRDIMM .. uh .. has made a lot of noise about the fact that they actually WROTE the LRDIMM spec .. uh .. but IDT is actually the only company today that has a JEDEC compliant LRDIMM product.

So we are very confident that we are going to .. uh .. continue to expand our share in LRDIMM just as we did in DDR3 and DDR2 and AMD before that.

This is taken from the article:

https://ddr3memory.wordpress.com/2012/05/24/lrdimm-buffer-chipset-makers/
LRDIMM buffer chipset makers
May 24, 2012

However, over the course of a few conference calls, IDTI has deemphasized LRDIMMs for Romley. And now they have completely skipped the LRDIMM Romley rollout.

This is a major strategic decision since by doing so they are foregoing early relationships with partners and OEMs.

IDTI has said they will target the post-Romley Ivy Bridge at 1600MHz (I assume at 3 DPC) by late 2012, and will try to catch the latter part of Romley if they can (which maybe hard to do since Inphi will have developed all the relationships).

In response to this behavior, Inphi has said they have a 9 months head start over their LRDIMM competitors (IDTI):

http://investors.inphi.com/phoenix.zhtml?c=237726&p=irol-irhome
Webcast Image Webcast
Q1 2012 Inphi Corp Earnings Conference Call (Replay)
04/24/12 at 2:00 p.m. PT

at the 37:10 minute mark ..

Doug Cripenwaithe (?) – RBC Capital Markets:

And then how about the competitive position.

Where do you see the .. uh .. have you started to see competition start to show up in the LRDIMM market or ..

at the 37:20 minute mark ..

Ford Tamer – CEO:

Yeah, I mean on the LRDIMM .. uh .. we do expect .. uh .. the LRDIMM to be multi-sourced (here only referring to other LRDIMM suppliers, like IDTI LRDIMMs, and NOT mentioning NLST HCDIMM as a competitor).

So .. uh .. we do expect competitors to be qualified.

As of right now, we are the only qualified solution, but do expect others to come in and be qualified.

The one .. uh .. I feel good about where we are, because we are 9 months ahead (of IDTI LRDIMMs etc.) of the next .. uh .. next competitor being qualified and that 9 months is very critical in the learning experience – understanding what works, what doesn’t work, what needs to be added.

How to qualify and it’s not a simple qualification process.

With different processor vendors (Intel, AMD), different module vendors (those who use Inphi LRDIMM “iMB” buffer chipset to make a memory module), different server OEMs, different application software.

Uh .. you know a pretty stringent number of .. uh .. tests .. uh .. from system to bench to .. uh .. component to ATE (?) to hot calls (?).

I think we .. we feel really good about the fact that we are ahead in understanding .. uh .. the requirements, the benefits, working with thend-users in both Fortune 500 and mega data centers to really understand what is gonna make a difference in a next-generation offering.

So do we expect other competitors – sure.

Do we expect our 9 months lead to give us benefits in the future – absolutely (Note: NLST HCDIMM has a greater lead than LRDIMM in this regard – as HCDIMMs already listed on VMware as only qualified memory – other is Kingston, for example).

at the 18:40 minute mark:

John Edmunds – CFO:

I think the other thing to keep in mind is that as another competitor comes in, it helps to validate the market (note how IDTI the “other competitor has scaled back rhetoric on LRDIMM and is now planning a post-Romley Ivy Bridge launch for it’s LRDIMMs).

And this is a JEDEC driven market, so you have end-users, or end system customers that WANT to see multiple suppliers.

And they are hesitant to get into a market where there aren’t multiplie suppliers (what if they also know that there is NLST vs. Inphi and on the back of NLST patent all claims surviving reexams – that a scenario where LRDIMMs are recalled is a possibility – how would that affect their confidence in the LRDIMMs JEDEC or no JEDEC).

Why did IDTI skip Romley rollout of LRDIMMs ?

Ostensibly the reason given is that LRDIMMs will not sell in volume with Romley.

This makes sense because 16GB LRDIMMs are non-viable.

However, why would IDTI not want to establish early relationship with partners and OEMs – before they get locked into one supplier (as memory module makers tend to do with buffer chipset makers).

Here is Inphi discussing the need to establish early relationships with memory module makers – since most settle on a single supplier (to minimize SKUs):

http://www.veracast.com/stifel/tech2011/main/player.cfm?eventName=2133_inphic
Stifel Nicolaus
Technology, Communications & Internet Conference 2011
Inphi Corporation
2/10/2011; 4:25 PM
Mr. John Edmunds
Chief Financial Officer

DISCLAIMER: please refer to the original conference call or transcript – only use the following as guidance to find the relevant section

at 24 minute mark ..

(comments on the competitive landscape)

in servers .. because of the qualification cycles. . there really are some incumbent competitors .. like IDT and TXN (Texas Instruments) .. and so the 3 of us tend to split the market.

IDT and Inphi would probably share 80% of the market. TXN would be somewhere in 10-15% range.

TI is not developing an LRDIMM to our knowledge and .. uh .. their interest level seems to wax and wane at times.

We go head to head with IDT – we respect them as competitors and we think market is going to want multiple suppliers.

It’s not a market that somebody from outside can come into easily just because of the long qualification cycles and the fact these are getting deployed across a wide range of SKUs (stock keeping units ?) .. the OEMs that (unintelligible) the memory module makers don’t want to qualify multiple suppliers because they have to deploy them across a wide set of SKUs ..

Comparing Inphi with IDTI

Inphi is a company seemingly without a retail shareholder base and which involves itself in litigation (with faulty patent portfolio – Inphi had to abandon the retaliatory Inphi vs. NLST for this reason possibly because of a case of double-patenting at the USPTO by Inphi). So Inphi behavior is far more aggressive than their capability – which is suggestive of them behaving as a proxy for other players (possibly their major institutional/company investors).

IDTI has a far less adverserial relationship with Netlist – and in fact is listed as a partner on the Netlist website (contrast this with Inphi).

With successful defence of IP at the USPTO (recent Netlist ‘537 and ‘274 patent reexam wins and failure of Inphi to challenge them at USPTO), it should now be clear that the Netlist IP is here to stay.

It is speculative, but it is possible that the far more prudent IDTI may have decided to avoid Romley rollout of LRDIMMs until the legal landscape was clear with regard to Netlist IP.

If so, it is reasonable to expect that IDTI would also want to license Netlist IP:

– license Netlist IP so they can produce LRDIMMs
.
– or become a buffer chipset second-source (similar to Diablo/Toshiba who were earlier designated as resellers of HyperCloud buffer chipsets for sale to third-parties by Netlist)

Second-source for HyperCloud

Netlist has been asked this question in a conference call – and they have said they can supply themselves but will have to have one in the long run.

Here is an analyst asking Netlist if there is pressure on them from the OEMs to enable a second-source:

http://seekingalpha.com/article/592411-netlist-s-ceo-discusses-q1-2012-results-earnings-call-transcript
Netlist’s CEO Discusses Q1 2012 Results – Earnings Call Transcript
May 15, 2012

at the 19:45 minute mark:

Rich Kugele – Needham & Co:

Thank you and good afternoon.

A couple of questions.

Uh first, when it comes to .. uh .. customers of such scale as HP and IBM .. uh .. obviously .. since they have signed contracts with you, they are comfortable with your ability to supply, but have they asked for anyone else to be a .. uh .. second-source in any way for these HyperCloud modules.

Have they asked you to license it out in any way.

at the 12:15 minute mark:

Chuck Hong – CEO:

Uh .. Rich .. uh ..

Uh .. we’ve had some of those discussions in the past.

I think as we see traction and the volumes increase over time .. uh .. we’ll get into more serious discussions about a potential second-source.

For now I think they are comfortable with us supporting .. uh .. product.

We’ve got products that have already been shipped into the hubs worldwide .. uh .. for both HP and IBM and .. uh .. we’ve started to .. uh .. fill the pipeline.

So I think we’ll discuss second-sourcing as the volumes increase.

See the section entitled “third-party manufacture of HyperCloud” in the article:

https://ddr3memory.wordpress.com/2012/07/03/examining-netlist/
Examining Netlist
July 3, 2012

A second source is especially important for Netlist because Netlist is both a buffer chipset maker (and holder of IP for load reduction and rank multiplication) – as well as a producer of the HyperCloud memory modules using those buffer chipsets.

There is a need for not only secondary memory module makers to appear in this space, but there should also be secondary suppliers of HyperCloud buffer chipsets.

If HyperCloud-based 16GB and more importantly 32GB (starting in late 2012 and ramping in 2013 and beyond) are to ramp up at the OEMs, they will require a second-source eventually in order to ensure supply (or to reduce risk of a supply interruption at the single-source supplier).

Time-to-market for second-source

Volumes for 32GB load reduction and rank multiplication product will start to increase starting late 2012 and early 2013 and be ramping thereafter.

Netlist will not have the time to establish a second-source then. Obviously some second-source will have to be arranged now, or would already have been arranged.

This is because anyone who licenses Netlist IP now and starts testing, producing and then qualifying at the OEMs will deliver mid-2013 or later – and would miss the chance to establish relationship with partners and OEMs for Ivy Bridge in late 2012.

I had posted in an earlier article that the likely third-party manufacture of HyperCloud may be timed for DDR4, and essentially skip DDR3 – because there would not be enough time to qualify those at the OEMs.

See the section entitled “third-party manufacture of HyperCloud” in the article:

https://ddr3memory.wordpress.com/2012/07/03/examining-netlist/
Examining Netlist
July 3, 2012

However, if someone already has a head start in integrating Netlist IP, they could still benefit from an Ivy Bridge rollout in late 2012.

This is a situation which seems to mirror the situation at IDTI – which lost interest in the highly important Romley rollout, and instead quietly moved deadlines to Ivy Bridge in late 2012.

IDTI as a licensee of HyperCloud IP for LRDIMMs

On the other hand, if IDTI were to simply negotiate a licensing deal – to provide legal cover for LRDIMMs – they would have the ability to produce a legally unencumbered LRDIMM for Ivy Bridge. And set themselves up for the 32GB rampup in late 2012 and 2013, in competition with Inphi (who would presumably not have legal cover).

But that would not explain the delay by IDTI in pushing LRDIMM.

IDTI rollout in late 2012 may be suggestive of a re-design at IDTI of the “LRDIMM” to better align it with DDR4. That is, it is possible IDTI could ship a more HyperCloud-like product than an LRDIMM product in late 2012.

IDTI as a second source for HyperCloud

IDTI’s sudden delay and targeting of a post-Romley Ivy Bridge in late 2012 fits the bill.

This is purely speculative, but if IDTI had this under wraps for a while i.e. they already have stable product working to deliver late 2012 – this would allow IDTI to participate in the 32GB market which will start ramping by late 2012 and into 2013.

As stated above, Inphi claims to have a 9 month advantage vs. IDTI. Inphi is basing that comparison on the “delay” at IDTI.

However, if IDTI have already been working on becoming a second-source of RDIMM-compatible HyperCloud – for delivery with Ivy Bridge – that would give them a one year advantage over Inphi (if Inphi were to start now with an RDIMM-compatible HyperCloud they could not deliver a similar product until mid-2013 at the OEMs).

Impact of availability of a RDIMM-compatible product

If IDTI were to market a RDIMM-compatible load reduction and rank multiplication product i.e. a “better RDIMM” (HyperCloud), that would solve another of the industry’s problems – LRDIMMs.

If RDIMM-compatible load reduction and rank multiplication products become available, many people may want to forget they ever heard of LRDIMMs – with all their attendant complications – a new standard, BIOS upgrade complications, latency issues and a design that does not translate over well to DDR4 (which copies the HyperCloud design even more).

Netlist has suggested that LRDIMMs are an “end-of-life” product.

See the section entitled “LRDIMMs – cul-de-sac for an “end-of-life” product” in the article:

https://ddr3memory.wordpress.com/2012/07/05/examining-lrdimms/
Examining LRDIMMs
July 5, 2012

Conclusion

If IDTI were to release an LRDIMM product for Ivy Bridge (with licensing from Netlist), it would be the first LRDIMM product with legal cover.

But what would happen if IDTI’s product for Ivy Bridge was not an LRDIMM product, but a RDIMM-compatible load reduction and rank multiplication product (HyperCloud).

If that happens, IDTI would be the first Netlist licensed LRDIMM/HyperCloud type of product in the market (without legal issues).

And would gain a one year or more advantage over Inphi – a surprising turn of events, because Inphi is currently claiming a 9 month advantage over IDTI.

Advertisements

1 Comment

Filed under Uncategorized

One response to “A second-source for HyperCloud ?

  1. Pingback: HyperCloud to own the 32GB market ? | ddr3memory

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s