Patent reexamination as a double-edged sword
UPDATE: added 06/24/2012: Invensas on LRDIMM design inferiority vs. HyperCloud
Some years ago MetaRAM arrived on the scene. It had the support of Intel (see link below) and some significant venture backing (Intel VC, Khosla Ventures).
DISCLAIMER: What follows is an examination of the efforts by companies to infringe on Netlist IP in the past. I am not a lawyer, nor am I qualified to give legal advice – what follows is my own understanding – readers should do their own research before they arrive at a conclusion.
Back in 2008, at the height of MetaRAM hype, the company was claiming the idea for their product was their own:
IDF: Intel gets behind start-up MetaRAM’s server memory solution
August 19, 2008 | Dean Takahashi | Comments |
It was founded in January 2006 by Suresh Rajan, who had a simple but elegant idea. He figured out a way to trick a server’s memory controller. Normally, a server can access a limited amount of memory chips on a single memory module. MetaRAM creates a chip set for the memory module that lets the memory controller count multiple chips as just a single one. In doing so, it lets the controller overcome the memory limitations and access as much as four times more memory on a module.
Virtualization – Start-Up Closes the Memory Gap; Cuts High-End Systems’ Prices 90%
If you heard a thump Monday, that was the cost of memory crashing
By Maureen O’Gara
March 4, 2008 03:15 PM EST
MetaRam’s secret is a memory chipset thought up by its co-founder ex-Nvidia/Rambus exec Suresh Rajan – and sorcery, of course.
Sorcery, no doubt.
The MetaRAM founder Suresh Rajan quoted above worked at Rambus prior to MetaRAM (and may have gone back to Rambus after MetaRAM closure).
Netlist vs. MetaRAM
As part of settlement in NLST vs. MetaRAM, MetaRAM conceded IP to NLST and said that they had destroyed all infringing product in court docs, and then MetaRAM went out of business.
July 8, 2009, 5:46 PM
Turning Out The Lights: Semiconductor Company MetaRAM
By Tomio Geron
MetaRAM used one patent (the ‘220 patent) in it’s retaliatory MetaRAM vs. Netlist lawsuit:
This patent was conceded to Netlist as part of settlement. As can be seen from the USPTO assignments page:
That patent is now licensed to NLST as part of settlement (and any buyer – Google or other – of this IP from MetaRAM will not be able to use it against NLST – see the language “protects Netlist if another company purchases MetaRAM’s patent” below).
From the PR at time of NLST/MetaRAM settlement:
Netlist Announces Settlement of Patent Infringement Lawsuits With MetaRAM
Press Release Source: Netlist, Inc. On Thursday January 28, 2010, 1:25 pm EST
Under the terms of the settlement, filed in U.S. District Courts in Delaware and Northern California, MetaRAM will not sell, offer to sell, release, or commercialize the MetaRAM DDR3 controllers in the U.S. or outside the U.S. Netlist contended that MetaRAM’s DDR3 controllers and memory modules incorporating such controllers infringed its U.S. Patent No. 7,289,386, entitled “Memory Module Decoder.” A provision in the settlement protects Netlist if another company purchases MetaRAM’s patent and attempts to seek action against Netlist in the future.
“We are pleased to have successfully resolved this case,” said C.K. Hong, President and CEO of Netlist. “As the pioneer of this technology, the results of this settlement clearly underscore Netlist’s fundamental patent and product leadership. Netlist’s HyperCloud product-line embodies this foundational technology and Netlist remains committed to protecting its portfolio of intellectual property.”
Despite the infringement of Netlist IP, there WERE some variations in how MetaRAM was producing it’s product (an emphasis on “stacking DRAM” ?) vs. Netlist’s way (which it said was better). Here is an example of such language from that time:
While some packaging companies stack devices to double capacity, Netlist achieves the same result without stacking, resulting in superior signal integrity and thermal efficiency. Stacking components results in unequal cooling of devices, causing one device to run slower than the other in the stack. This often results in module failures in high-density applications.
The density limitation is solved by proprietary board designs that use embedded passives to free up board real estate, permitting the assembly of more memory components on the substrate. The performance of the memory module is enhanced by fine-tuning the board design to minimize signal reflections, noise, and clock skews.
Netlist vs. Inphi (maker of LRDIMM buffer chipsets)
Along similar lines we have the Netlist vs. Inphi and the retaliatory Inphi vs. Netlist lawsuits:
Netlist Files Patent Infringement Lawsuit Against Inphi
On Wednesday September 23, 2009, 7:09 pm EDT
At the time, Inphi called it a “nuisance lawsuit” based on the ‘537 NLST patent (more on that below):
September 28, 2009 08:00 AM Eastern Daylight Time
Inphi Will Vigorously Defend Itself in Recent Nuisance Lawsuit
WESTLAKE VILLAGE, Calif.–(BUSINESS WIRE)–Inphi Corporation, a high-speed analog semiconductor company, announced today that it will vigorously defend itself in response to a nuisance lawsuit announced by Netlist, Inc. on September 23, 2009. The Netlist press release claims that Inphi’s newly developed iMB ™ technology infringes on Netlist’s U.S. Patent No. 7,532,537. While Inphi has not received any communication from Netlist on this matter, nor been served with a complaint, Inphi has examined the patent in question and believes there is no infringement.
Inphi’s iMB technology has been widely praised in the marketplace and is now gaining significant traction. Inphi intends to continue to develop innovative technologies in earnest and will respond accordingly to contrived efforts to obstruct our progress,” said Young Sohn, president and CEO of Inphi.
Inphi and MetaRAM
Some years after the demise of MetaRAM, we have Inphi hiring former MetaRAM CEO Fred Weber as “Technical Advisor” by Inphi – so it is perhaps not surprising that Inphi seems to be going down a similar road:
Inphi Taps Industry Visionaries As Technical Advisors
Experts Provide Unique, Global Insight in Technology, Strategy and Direction
WESTLAKE VILLAGE, Calif., February 23, 2010
“I’ve spent my career focused on building balanced computer systems and providing compatible and evolutionary innovations,” said Weber. “Inphi shares my world view that, with the emergence of multi-core and virtualization technologies, the memory system is once again the biggest bottleneck in systems. Inphi recently announced an isolation Memory Buffer component that addresses this dilemma by delivering a new memory scheme that multiplies the server’s memory capacity as much as fourfold. The Inphi team continues to raise the bar and I am happy to be part of their advisory team.”
Changing language in Inphi’s SEC filing about infringement
An earlier IPHI SEC filing has some risk factors and info on the lawsuits and the impact it may have on IPHI’s memory module (buffer chipsets) business:
INPHI Corp · 424B4 · On 4/1/11
For example, Netlist, Inc. filed suit against us in the United States District Court, Central District of California, in September 2009, alleging that our iMB ™ and certain other memory module components infringe three of Netlist’s patents. For more details, see “Business—Legal Proceedings.”
Infringement claims also could harm our relationships with our customers or distributors and might deter future customers from doing business with us. We do not know whether we will prevail in these proceedings given the complex technical issues and inherent uncertainties in intellectual property litigation. If any pending or future proceedings result in an adverse outcome, we could be required to:
* cease the manufacture, use or sale of the infringing products, processes or technology;
* pay substantial damages for infringement;
* expend significant resources to develop non-infringing products, processes or technology, which may not be successful;
* license technology from the third-party claiming infringement, which license may not be available on commercially reasonable terms, or at all;
* cross-license our technology to a competitor to resolve an infringement claim, which could weaken our ability to compete with that competitor; or pay substantial damages to our customers or end users to discontinue their use of or to replace infringing technology sold to them with non-infringing technology, if available.
Any of the foregoing results could have a material adverse effect on our business, financial condition and results of operations.
Following the Inphi loss at the USPTO with Netlist ‘537 and ‘274 patents surviving reexamination with ALL claims intact, one would think these risk factors would still apply.
Yet, they have paradoxically vanished from the most recent Inphi SEC filings:
INPHI Corp · 10-Q · For 3/31/12
Filed On 5/9/12 2:32pm ET · SEC File 1-34942 · Accession Number 1193125-12-221626
Evidently Inphi now thinks it no longer needs to warn shareholders about the risks of infringement.
Patent reexamination as a double-edged sword
However, Inphi employed a dilatory strategy – leveraging the delaying power of patent reexaminations at the USPTO to have Netlist vs. Inphi stayed at the courts.
Patent reexamination is a long drawn out process where a patent owner’s patents are reexamined “from scratch” at the USPTO (i.e. there is “no presumption of validity” when examining the patent afresh during reexams):
THE NEW ROLE OF REEXAMINATION IN PATENT LITIGATION
(pg. 16 ):
An explicit intent of the reexamination procedures is “to maximize respect for the reexamined patent.” 54 Although not explained, this statement does not infer any “presumption of validity” to the patent being reexamined. In reexamination, “there is no presumption of validity and the ‘focus’ of the reexamination ‘returns essentially to that present in an initial examination.’” 55 Reexamination by definition is just that, and requires the examiner to apply the same analysis as for an original examination.
While the patents remain valid during the reexam process, for the purposes of the patent reexamination, the patent is built up from scratch by first assuming it is invalid and then the patent is slowly built up as the challengers arguments and patent owner’s arguments are heard.
The process is similar to when the patent was first awarded, except now the challenger is given the opportunity to stop development and re-assembly of the patent at every step.
The benefit of this (for the challenger) is that by employing this tactic, they are able to:
– delay litigation (since courts dealing with cases which are still in the early stages tend to rule for stay of proceedings pending resolution at the USPTO i.e. “why not wait a bit”)
– the potential exists for all claims of the patent to be rejected (and this is esp. true of patents which are not vigorously defended by the patent owner)
However, there is a danger for the challenger as well (the double-edged sword) – if some claims survive the reexamination process:
– the surviving claims become IMPOSSIBLE to challenge
– a court will not even allow a challenger to hint those patent claims could be invalid
This means that patent reexaamination may be a good dilatory strategy (for example, a delay in court proceedings may allow the company to go about it’s business, engage in an IPO without fear of interruption).
However the downside is that IF the patent reexamination leads to any claims surviving reexam, those claims become impossible to refute any further.
To clarify, a “third-party requester” is the party who is challenging a patent:
THE NEW ROLE OF REEXAMINATION IN PATENT LITIGATION
(pg. 3 ):
But inter partes reexamination has not been pursued routinely for fear it will backfire with an affirmation of the patent claims by the Patent Office, and create estoppel in the litigation as to any ground or issues the requester raised or “could have raised.” 4
4 An estoppel adverse to a third-party requester (which does not exist in ex parte reexamination) will attach in the case of an inter partes reexamination, if the requester is unsuccessful in the inter partes reexamination proceeding. The requester is estopped from later asserting in any civil action, or in a subsequent inter partes reexamination, the invalidity of any claim finally determined to be valid and patentable on any ground the third-party requester raised or could have raised in the inter partes reexamination. 35 U.S.C. § 315(c) (2006). Also, the requester might be estopped from later challenging in a civil action any “fact” determined in the inter partes reexamination.
As it turned out, the USPTO completed their reexamination of the NLST ‘537 and ‘274 patents with all claims surviving reexamination – a catastrophic outcome for Inphi. This is a powerful signal of the strength of NLST IP. Inphi still would like the BPAI (Board of Appeals and Interferences) to take a look at it, but it is clear that a complete reversal (because Inphi would need to invalidate each and every claim of the patent) looks unlikely.
Netlist Provides Update on US Patent & Trademark Office Communications
USPTO Allows All Claims in the Reexamination of Netlist’s ‘537 and ‘274 Patents
MarketwirePress Release: Netlist, inc. – Tue, Mar 27, 2012 4:05 PM EDT
IRVINE, CA–(Marketwire -03/27/12)- Netlist, Inc. (NASDAQ: NLST – News), a leading provider of high performance memory solutions for the cloud computing market, today announced that the United States Patent and Trademark Office (“USPTO”) recently issued two favorable Office Actions in the reexaminations of US Pat. No. 7,532,537 (‘537) and US Pat. No. 7,636,274 (‘274). In the reexamination of the ‘537 patent, the USPTO allowed all 60 claims, while in the reexamination of the ‘274 patent, the USPTO allowed all 104 claims.
By allowing these claims over the cited art, the USPTO rejected each and every prior art challenge raised by Inphi Corporation in its initial reexamination requests filed in June 2010, and subsequently during the reexamination proceedings of the ‘537 patent and the ‘274 patent. Inphi has a right to appeal these decisions.
“The USPTO’s current allowance of these claims solidifies Netlist’s leadership position in high-performance logic-based memory solutions that utilize the innovations claimed in the ‘274 and ‘537 patents,” said Dr. Hyun Lee, Vice President, Chief Technology Officer of Netlist. “This clearly points to the significant role that Netlist can play in developing cutting-edge next generation products using current and future DDR memory technologies.”
Netlist’s HyperCloud™ HCDIMM™ is one example of Netlist’s currently available products that deliver superior performance and are based in part on the innovative concepts protected by the ‘537 and ‘274 patents. HCDIMM has been proven to substantially outperform Load Reduced DIMMs (“LRDIMM”) when benchmarked on the latest generation Intel® Xeon® Processor E5-2600 motherboard configured with 384GB system memory. HCDIMM’s innovative architecture results in superior memory performance compared to LRDIMM’s monolithic memory buffer implementation. Information on HyperCloud’s industry leading performance, including recent independent tests and OEM qualification can be found at http://www.netlist.com.
Inphi vs. Netlist (retaliatory)
Meanwhile Inphi has dropped it’s retaliatory lawsuit i.e. Inphi vs. Netlist.
If Inphi had pursued Inphi vs. Netlist, there may have been a danger that one or both of those Inphi patents may have been invalidated.
United States District Court Orders Dismissal of Inphi Versus Netlist Patent Infringement Lawsuit
Press Release Source: Netlist, Inc. On Thursday April 21, 2011, 6:00 am EDT
IRVINE, Calif., April 21, 2011 /PRNewswire/ — Netlist, Inc. (NASDAQ:NLST – News), a designer and manufacturer of high-performance memory subsystems, today said the patent infringement lawsuit filed by Santa Clara, CA-based Inphi Corporation, a high speed analog semiconductor company, against Netlist has been dismissed without prejudice. The lawsuit filed on November 30, 2009, in the U.S. Central District of California, Western Division followed a patent infringement lawsuit Netlist filed against Inphi on September 22, 2009.
“We are pleased with the court’s order to dismiss Inphi’s complaint against Netlist,” said C.K. Hong, President and CEO of Netlist. “We believe our robust portfolio of intellectual property uniquely addresses the memory capacity and performance challenges our customers face in cloud computing, virtualization and high performance computing. We will continue to aggressively defend our patent rights.”
Netlist’s ongoing federal litigation against Inphi Corporation alleges that Inphi’s iMB line of isolation memory buffer integrated circuits infringe Netlist’s U.S. Patent Nos. 7,532,537, 7,636,274 and 7,619,912. Netlist is seeking unspecified monetary damages and has requested an injunction barring Inphi from infringing and inducing others to infringe any of the claims of the ‘537, ‘274 and ‘912 patents.
Reasons for retraction of Inphi vs. Netlist (retaliatory)
From my reading of court docs, I suspect the reason maybe that the two patents Inphi was using to prosecute this case were pointed out by Netlist to be suffering from a case of “double patenting” – the same patent was submitted by two different set of authors to two different examiners – in what looks like a case of patent inflation !
These are the two Inphi patents which it was using in Inphi vs. Netlist:
DISCLAIMER: I am not a lawyer, nor am I qualified to give legal advice – the above is my reading of the patent documents. Readers should examine the patent documents themselves to arrive at their own conclusions.
Here is how this was described in Inphi SEC filings:
INPHI Corp · 424B4 · On 4/1/11
For example, we filed a complaint against Netlist in Federal District Court in November 2009 alleging that Netlist infringes two of our patents. Netlist asserts in its amended answer to the complaint that it does not infringe the patents, that the patents are invalid and that one of the patents is unenforceable due to inequitable conduct before the United States Patent and Trademark Office, or the USPTO. For more details, see “Business—Legal Proceedings.”
Inphi’s “intent to deceive” the USPTO
In docket 18 (Inphi vs. NLST), NLST claims that although the same IPHI Attorney Patrick filed both the ‘863 and the newer ‘799 patents (which was incidentally filed under DIFFERENT author names), the prior ‘863 patent was not mentioned in the prosecution of the ‘799 patent. While the ‘799 patent is substantially similar to the earlier ‘863 patent. In addition the USPTO examiners for the two patents were different.
Since parties are required to disclose similar patents to prevent dual-patenting (and to artificially extend the life of patents), NLST contends this was done “with an intent to deceive” the USPTO.
Ordinarily you would file a patent and any patents which are “prior art”. If a new patent is very similar to an earlier patent, the USPTO may require the patent owner to agree to a “Terminal Disclaimer” before the patent is awarded. This is to ensure that the “new patent” does not have a life beyond the life of the previous patent. We can see that in the case of some of NLST’s patents – for example the ‘150 patent.
At the very least the ‘799 patent is unpatentable (in NLST view) for these reasons.
Further, NLST claims that because of the irregularities in the ‘799 patent process, the ‘863 patent (earlier one) is also invalid under the “doctrine of equivalents”:
In the United States, the doctrine of equivalents is limited by prosecution history estoppel.
This presumptive bar approach holds that where claims are amended, “the inventor is deemed to concede that the patent does not extend as far as the original claim” and the patentee has the burden of showing that the amendment does not surrender the particular equivalent. To succeed, then, the patentee must establish that:
– the equivalent was unforeseeable at the time the claim was drafted;
– the amendment did not surrender the particular equivalent in question; or
– there was some reason why the patentee could not have recited the equivalent in the claim.
Essentially because the ‘799 patent prosecution did not mention the ‘863 patent that implicitly means that IPHI is surrendering the rights under the ‘863 patent.
It is for these reasons, that it seems possible that Inphi unilaterally retracted Inphi vs. Netlist (retaliatory) because further pursuit would have invalidated one or both of the patents it was using against Netlist, and may have been the cause of considerable embarrassment for Inphi before the USPTO.
Which leaves us with Netlist vs. Inphi
Netlist vs. Inphi was stayed by the judge pending patent reexamination of NLST patents at the USPTO.
Once NLST patents emerge from USPTO having survived reexamination in whole or in part, they will not be challengeable by Inphi in court.
Which does not bode well for Inphi in court.
Impact on LRDIMMs buffer chipsets
LRDIMM buffer chipsets are made by the buffer chipset makers (Inphi) and are used by memory module makers (Samsung, Hynix) to make LRDIMM memory modules.
Of the top 3 buffer chipset makers:
– IDTI has prudently scaled back on the rhetoric over the course of a few quarters (as evidenced by their enthusiasm for LRDIMMs during conference call comments).
– IDTI has postponed LRDIMMs to end of 2012 (i.e. skipping Romley and targeting the Ivy Bridge series according to their conference call).
– Texas Instruments has not been interested in LRDIMMs – possibly related to settlement in Netlist vs. Texas Instruments a couple of years ago (it seems Texas Instruments may have been the original leaker of NLST IP to JEDEC).
– Only Inphi has persisted with LRDIMMs and are currently the only supplier of buffer chipsets for LRDIMMs.
Inphi has also been the most aggressive of the three – having challenged NLST’s right to their IP in patent reexamination challenges at the USPTO.
However, these challenges have led to the strengthening of Netlist’s position – as patents which survive reexamination become extremely difficult to refute in court.
Impact of litigation on LRDIMMs and DDR4
LRDIMMs are currently underperforming the competition (RDIMMs in the 1 DPC, 2 DPC niche and HyperCloud in the 3 DPC niche (at 16GB) and 3 DPC and 2 DPC (at 32GB) as explained in other articles here).
Which may make issues about litigation moot – since there may be no “infringing product”.
However, the impact may be more significant for DDR4 – since it incorporates ideas from LRDIMMs, but has dumped some of the asymmetrical lines and centralized buffer chipset choices in the LRDIMMs, in favor of an even closer following of the symmetrical lines and decentralized buffer chipset on the Netlist HyperCloud:
Netlist puffs HyperCloud DDR3 memory to 32GB
DDR4 spec copies homework
By Timothy Prickett Morgan
Posted in Servers, 30th November 2011 20:51 GMT
Here is a CMTL labs comparison of HCDIMM and LRDIMM that goes into more detail:
CMTL HCDIMM Outperforms LRDIMM in “Big Data” & “Big Memory” Applications White Paper
UPDATE: added 06/24/2012: Invensas on LRDIMM design inferiority vs. HyperCloud
Here is a paper that describes a future DDP (dual die packaging) design. One of the authors – Bill Gervasi – is a former Netlist employee and a former JEDEC committee chair. The paper mentions the applicability to both LRDIMMs and HyperCloud. Here is the section where they compare the design weaknesses in the LRDIMMs with the superiority of the HyperCloud design:
Cost-minimized Double Die DRAM Packaging for Ultra-High Performance DDR3 and DDR4 Multi-Rank Server DIMMs
Richard Crisp 1 , Bill Gervasi 2 , Wael Zohni 1 , Bel Haba 1
1 Invensas Corp, 2702 Orchard Parkway, San Jose, CA USA
2 Discobolus Designs, 22 Foliate Way, Ladera Ranch, CA USA
5. Applicability to LRDIMM and Hypercloud DIMMs
The LRDIMM differs from the RDIMM in that the DQ and DQ Strobe signals are buffered. The data buffer is placed in the central region of the LRDIMM. This requires all data and data strobes to be routed from each DRAM package to the buffer and then routed back to the edge connector which demands additional routing layers versus an RDIMM. Since the LRDIMM is plugged into an edge connector, the thickness of the DIMM PCB is fixed.
Adding PCB layers necessarily requires a reduction of the thickness of the dielectric layers separating the power planes and routing layers. Unless the width of the traces is made narrower, the characteristic impedance of the etched traces is decreased and can lead to signal reflections arising from impedance discontinuities that diminish voltage and timing margin.
Trace width is limited by the precision of the control of the etching process, with such narrower traces being more costly to manufacture within tolerance. Because the DFD’s C/A bus routes on a single layer and other interconnections lay out cleanly, the layer count is reduced leading to nominal impedances being attainable with normal dimensional control keeping raw card costs from rising.
The Hypercloud architecture is similar to the LRDIMM in that the DQ and DQ Strobe signals are buffered, but unlike the LRDIMM the buffering is provided by a number of data buffer devices placed between the edge connector and the DRAM package array on the DIMM PCB. The 11.5 x 11.5 mm package outline of the DFD supports placement of the buffers without requiring growth of the vertical height of the DIMM. In fact a simple modification of the RDIMM PCB will enable the Hypercloud data buffers to be mounted on the PCB making conversion of an RDIMM design to Hypercloud a straightforward matter.
Here Invensas is saying that with the LRDIMM centralized buffer chipset, there are a greater number of lines to route back and forth between DRAM packages and the central buffer. This forces the use of a greater number of PCB layers, each of which then needs to be thinner, and leads to signal quality issues.
DDR4 intersection of Netlist IP
While DDR4 is often portrayed as borrowing a few ideas from LRDIMMs, in fact LRDIMMs (with their asymmetrical lines and centralized buffer chipset) are an “end-of-life” product.
Netlist has said for a while that LRDIMMs will have latency issues because of their asymmetrical lines and centralized buffer chipset (something which was only confirmed in early 2012 as information about LRDIMMs started becoming known – mainly through the Inphi LRDIMM blog).
Netlist has also indicated that DDR4 will intersect Netlist IP:
Fourth Quarter and Full Year 2011 Conference Call
Tuesday, February 28 5:00pm ET
at the 15:45 minute mark ..
Now that we are qualified, we will work directly with the OEMs to introduce our products to end-users who will then run the product in their particular applications.
This process has already begun with some end-customers.
at the 16:00 minute mark ..
After this, we expect them to start placing orders.
Once adopted by these end-users, and ultimately deployed in a variety of applications, we believe HyperCloud will become the standard high-end memory configuration and will be replicated as data centers scale.
at the 16:20 minute mark ..
We are now in the process of qualifying 32GB HyperCloud at the major server OEMs, and we expect the testing to completed by mid-year.
at the 16:30 minute mark ..
This product is also being evaluated by major storage and appliance manufacturers.
at the 16:35 minute mark ..
For the Romley platform launching in March (2012), as well as the addition of 32GB to the product line, we expect to see an expansion of HyperCloud revenues in 2012 – including substantial growth in the second half of the year, as we start shipping in volume.
Just as important, we believe that HyperCloud’s architectural advantages will extend through the end of the decade with DDR4 as server memory speeds go from 1333MHz currently to 2.5GHz at DDR4, while existing memory technologies continue to lag behind, the current gap between CPU and memory will widen even more.
at the 17:15 minute mark ..
In such an environment, the value of HyperCloud and associated IP become increasingly valuable.
While NLST has worked on this technology for almost 8 years, the advantages of HyperCloud’s distributed architecture is just beginning to be understood by the rest of the industry.
at the 17:35 minute mark ..
As such, the industry is presently embarking on a DDR4 server memory standard, which employs our distributed architecture, trying to accomplish a DDR4 server memory technology which exists today at DDR3 in HyperCloud.
at the 14:50 minute mark ..
While HyperCloud’s technology will scale to DDR4 and beyond, by the contrast the current monolithic architecture of LRDIMM will go end-of-life with DDR3.
All of this points to HyperCloud being a technology which is a generation ahead today, while possessing the horsepower to remain in the lead through DDR4.
First proprietary memory to gain wide-spread industry adoption
From NLST’s Q1 2012 CC:
Netlist’s CEO Discusses Q1 2012 Results – Earnings Call Transcript
May 15, 2012
With today’s joint announcement with HP, HyperCloud is now available for sale on HP’s top-selling newly-released Gen8 servers. HyperCloud has also been designated as an HP Smart memory. This coupled with availability on IBM System x servers makes HyperCloud the first ever proprietary memory technology to be widely adopted by the mainstream server market. HP and IBM account for a majority share of the server market and they combine for even a larger share of the world’s high-density server memory consumption.