Global Lambda Integrated Facility

Subject Re: The Fate of Lambda Rail: He Who Has the Gold Makes the Rules
From Gordon Cook <cook@xxxxxxxxxxxxxx>
Date Sun, 4 Dec 2011 21:20:59 -0500

The Fate of Lambda Rail: He Who Has the Gold Makes the Rules

However, in line with the above hope, my attendance at the Supercomputing 11 meeting in Seattle in mid November was not encouraging.  In mid summer I asked about NLR’s attendance at SC11.  I was told that it would surely be as large or larger that New Orleans SC2010.  When I visited the show floor I could not find an NLR booth of any sort.  The University of Tokyo (see photos this page) had a sign in its booth mentioning an affiliation with NLR.  Other than that nothing.  

Now the increasingly ubiquitous Jen Hodson did put out a press release saying that Patrick’s NLR was bring 100 gigabits per second connectivity via SCINet to the show floor.  We read “NLR is able to provide these services in partnership with Cisco, whose ASR 9000 Series router system and ONS 15454 optical transport platform power the 100G network. Cisco supplied the networking equipment in support of NOAA’s activities at SC11.”  

We may safely assume that Cisco bore the cost of having its optical gear highlighted in connecting bandwidth from NLR to the show floor and the the NOAA booth since, in a second press release from Jen on the very next day, we read “We are very delighted to have Cisco as our corporate partner as we build together the new infrastructure for 21st century, personalized medicine, “ said Dr. Patrick Soon-Shiong, Chairman of NLR and the CSS Institute.”  So Cisco in effect gets the contract for a new generation of 100 gig optical gear.  The value of the contract is certainly in the amount of a few tens of millions of dollars.  “The new routing and switching technology provided through the Cisco ASR 9000 platform positions NLR to develop the most scalable, highly available and high performance services required to handle the wide range of applications and content in the rapidly expanding NLR customer base” [Editor: The Cisco ASR 9000 series is carrier Ethernet. In the photo to the left it is contained with Ciena equipment in the cabinet identified as “Cisco Ciena.” 

In line with the increasing privatization of what used to be public resources -- King Abdulah University of Science and technolgy in Saudi  Arabia, shown in photos again to the left, had one of the most impressive booths at SC11. 

Meanwhile as a part of the NLR deal, Cisco also gets to keep the use of NLR as a demonstration show place for its telepresence systems.  

For its part NLR, in my opinion, is not getting the most up to date and efficient optical equipment.  That is not Cisco but rather the Nortel equipment used by SURFNet two years ago and awarded to Ciena this year for SURFNet Seven.  (Ciena has acquired the Nortel patents and now makes the industry’s leading gear. )  Cisco had the contract for SURFNet 5 but lost it favor or Nortel’s far more cost effective lambda switching for SURFNet 6.

The November 16th release states that “it was announced in July that Dr. Patrick Soon-Shiong and the Chan Soon-Shiong Institute for Advanced Health had assumed financial responsibility for NLR, with a funding commitment of over $100 million.”  Two sentences later we read: “This enhanced infrastructure is now being put in place, transforming NLR from only a communications network into a translational healthcare ecosystem, to realize the promise of genomic medicine and bring the power of supercomputing to the point of care. It will also serve as a platform for innovative services and transformation in education, science and other areas.”

I believe that, at this point, Dr Soon Shiong clearly does not understand the reality with which he is dealing.  I really doubt that Tom de Fanti or Larry Smarr or Ron Jonson would agree that their last five years of research on NLR has been research “on only a communications network!”  On the basis of my own research in this area over the past three years I would say that for this full period of  time, it has “served as as a platform for innovative services and transformation in education, science and other areas.”  Past tense - not future.

Furthermore by not attending SC11, the new management of NLR seemed to put NLR behind a black out wall and thereby deprived researchers of the ability to talk with NLR people directly at SC 11. Perhaps inadvertently, Patrick has created a situation of uncertainty where researchers who have to plan their grant funding well in advance will have nothing to do except to gravitate except Internet2 which has its own problems of insularity.  Patrick seems to have pushed the old NLR Board aside and replaced it with an advisory council of his own choosing. This is unlikely to please a number of powerful people at their respective universities.  (The Board listed on the NLR site has a 2010 copyright. I have been unsuccessful in determining its 2011 role.  My requests for information from NLR have been ignored.)

I had an extensive conversation at SC11 with the head of a major state level optical research network who said his state had left NLR because he couldn’t afford to pay for membership in both NLR and I2.  He maintained that several other RONs had done likewise because a cash strapped NLR had been raising its rates. Whatever one’s interpretation, when you visit the NLR website you are greeted with an opaque exterior.  In 2010 the site had 23 press releases. in 2011 - ONE the July 27th acquisition.  Jen Hodson is listed as the media contact but her releases appear on Business Wire something that since NLR is now the flagship entity for Dr Soon Shiong’s  investment interests is not surprising.  Almost all the NLR pages have not been updated in the year 2011.

Finally in yet another November 16, 2011 press release (fittingly on the “I-Stock Analyst” site) we see that the CSS Institute announced in March that it was developing two data centers in Phoenix and Scottsdale, dedicated to health information, with more data center development planned near Sky Harbor airport. The first two data centers are now operational. The CSS Institute has also taken delivery of a unique, high performance computing facility, custom designed to perform rapid whole genome analysis. This “supercomputer” is now in Phoenix and operational. The CSS Institute plans further development in Phoenix, and will work closely with both Arizona State University and the University of Arizona. An affiliated body funded by Dr. Soon-Shiong, the Healthcare Transformation Institute, led by former Mayo Clinic CEO Dr. Denis Cortese, was established in Phoenix in partnership with both universities in 2010.

“Phoenix is becoming a healthcare hub,” said Dr. Soon-Shiong, who also heads National LambdaRail (NLR), a 12,000 mile high performance optical fiber network linking hundreds of research institutions across the United States. He explained that the CSS Institute’s data centers and supercomputer are now linked to NLR, and would soon be connected via NLR with genomic sequencing centers, medical researchers and healthcare practitioners throughout the country. Phoenix would therefore be a transportation hub for large genomic health data sets, as well as a center for data storage and genomic analysis, all linked to hospitals and physicians to support clinical decision-making, thereby for the first time bringing the promise of genomic medicine, and the power of supercomputing to the point of care.” [Editor: the most significant basic grid software Globus is now available on line for free.  See photo above.]

The context: “Mayor Phil Gordon will join the CSS Institute when he steps down after eight years as Mayor of the nation’s sixth largest city at the end of this year. Mayor Gordon will lead the CSS Institute’s growing interests in Arizona.”  Patrick is then quoted as saying: “So Phoenix is really in a pivotal position,” he continued. “Mayor Gordon has been a strong promoter of the life sciences, and I could not think of a better person to join us in the non-profit sector to help build a transformative national healthcare infrastructure in which Phoenix will play a crucial role.”

“I have always said that being Mayor of Phoenix is the best job in the world, but Dr. Soon-Shiong’s work may well be the most important,” said Mayor Phil Gordon. “I am convinced that, because of the technology infrastructure he is committing to our future, Phoenix will soon be to 21st century health what Silicon Valley is to the internet. I look forward to continuing my service to the future of our community in this new and wonderful way.

COOK Report: Yet one more time that the public servant retires to the employment by private sector interests.  

With reference to the statement above “that the CSS Institute’s data centers and supercomputer are now linked to NLR,” we also find out that on November 9, 2011 Patrick’s interests had signed a deal with a new privately held data center company named IO - headquartered in Phoenix.  IO’s press release announced that IO “has been awarded a contract with the Chan Soon-Shiong Institute for Advanced Health (CSS Institute), NantWorks LLC and National LambdaRail (NLR), to support the largest health-care data center in the world. As part of the agreement, IO will provide private cloud computing services at IO Phoenix.”

"Now, for the first time, we are able to bring together a high-performance communications network, dedicated storage capacity and high-performance computing," said Dr. Patrick Soon-Shiong, Chairman of the CSS Institute, NantWorks and the NLR. "This is the infrastructure that we have so far lacked for healthcare, and which is vital to translate genomic data into actionable clinical decision making. It is the beginning of 21st century medicine." “When completed, the central health-care database will be used to securely store patient information, to be accessible nationally and eventually globally. IO’s highly secure and reliable data center infrastructure enables all sectors involved in the healthcare industry such as research organizations, physician groups, hospitals, third-party payers and patients to access releasable data more quickly to help make research or clinical decisions. The database will have connection to National LambdaRail, a 12,000 high-performance communications network linking major research institutions throughout the United States.”

So there is a significant answer to the infrastructure problems posed at the beginning of this study.  The network is in place, the supercomputer has been installed.  From Business wire: “CSS Institute has also taken delivery of a unique, high performance computing facility, custom designed to perform rapid whole genome analysis. This "supercomputer" is now in Phoenix and operational.” (I have not yet ascertained which machine he bought - it may well be custom made.) But readers should note that genome sequencing is already big business.  Super computers devoted to this are already in place in Texas, Minnesota and Pittsburg.  At genomeweb we find in a May 9, 2011 announcement that has a podcast by Aleksei Aksimentiev, an assistant professor at the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign, describing his approach to personalized medicine with supercomputers. . . . . Aksimentiev is using the Ranger supercomputer at the Texas Advanced Computing Center to develop cheap DNA sequencing using nanopores.”  See also from Oak Ridge labs.  In an October 2011 talk at Pittsburg Craig Venter pointed out that “the human genome can be sequenced in less than two weeks for $4,000 using supercomputers the size of less than 3 feet by 3 feet and containing powerful 1.5 terabyte chips.”

On November 30, 2011 an article in the New York Times pointed out that that genome sequencing has become a huge and exploding field without adequate infrastructure to store and transmit the data. “BGI, based in China, is the world’s largest genomics research institute, with 167 DNA sequencers producing the equivalent of 2,000 human genomes a day.  BGI churns out so much data that it often cannot transmit its results to clients or collaborators over the Internet or other communications lines because that would take weeks. Instead, it sends computer disks containing the data, via FedEx. . . .

The result is that the ability to determine DNA sequences is starting to outrun the ability of researchers to store, transmit and especially to analyze the data.

“Data handling is now the bottleneck,” said David Haussler, director of the center for biomolecular science and engineering at the University of California, Santa Cruz. “It costs more to analyze a genome than to sequence a genome.”

That could delay the day when DNA sequencing is routinely used in medicine. In only a year or two, the cost of determining a person’s complete DNA blueprint is expected to fall below $1,000. But that long-awaited threshold excludes the cost of making sense of that data, which is becoming a bigger part of the total cost as sequencing costs themselves decline.

“The real cost in the sequencing is more than just running the sequencing machine,” said Mark Gerstein, professor of biomedical informatics at Yale. “And now that is becoming more apparent.”

But the data challenges are also creating opportunities. There is demand for people trained in bioinformatics, the convergence of biology and computing. Numerous bioinformatics companies, like SoftGenetics, DNAStar, DNAnexus and NextBio, have sprung up to offer software and services to help analyze the data. .  .  .  .

The lower cost, along with increasing speed, has led to a huge increase in how much sequencing data is being produced. World capacity is now 13 quadrillion DNA bases a year, an amount that would fill a stack of DVDs two miles high, according to Michael Schatz, assistant professor of quantitative biology at the Cold Spring Harbor Laboratory on Long Island.

There will probably be 30,000 human genomes sequenced by the end of this year, up from a handful a few years ago, according to the journal Nature. And that number will rise to millions in a few years.

In a few cases, human genomes are being sequenced to help diagnose mysterious rare diseases and treat patients. But most are being sequenced as part of studies. The federally financed Cancer Genome Atlas, for instance, is sequencing the genomes of thousands of tumors and of healthy tissue from the same people, looking for genetic causes of cancer.

COOK ReportThere is clearly opportunity here but an unbiased assessment is difficult to find.  

“We have these giant piles of data and no way to connect them” said H. Steven Wiley, a biologist at the Pacific Northwest National Laboratory. He added, “I’m sitting in front of a pile of data that we’ve been trying to analyze for the last year and a half.”

Still, many say the situation will be manageable. Jay Flatley, chief executive of Illumina, the leading supplier of sequencing machines, said he did not think information handling was a bottleneck or that it was causing people to hold off on buying new sequencers.

Researchers are increasingly turning to cloud computing so they do not have to buy so many of their own computers and disk drives.”  .  .  .  .  . “Google has enough capacity to do all of genomics in a day,” said Dr. Schatz of Cold Spring Harbor, who is trying to apply Google’s techniques to genomics data. Prodded by Senator Charles E. Schumer, Democrat of New York, Google is exploring cooperation with Cold Spring Harbor.

Google’s venture capital arm recently invested in DNAnexus, a bioinformatics company. DNAnexus and Google plan to host their own copy of the federal sequence archive that had once looked as if it might be closed.”

COOK Report: Dr. Soon Shiong has a lot of competition. And with the hole blown in the economy by Wall Street, the pressure is clearly there to privatize what was being done for the public good.

Extreme Science and Engineering Discovery Environment or XSEDE -- A New National Grid Independent of Dr Soon Shiong’s Phoenix Infrastructure

COOK Report: Furthermore it is also fortunate that private wealth has not yet been able to dislodge the role of the National Science Foundation in building infrastructure.  A new five-year program, called Extreme Science and Engineering Discovery Environment, or XSEDE, will replace the TeraGrid linkage that now connects the country's supercomputing resources. NICS is partnering with the University of Illinois, the lead institution on XSEDE. 

"The XSEDE grid will allow scientists and engineers to create scientific workflows and gateways, analyze data, and model complex phenomena," Blacklight at the Pittsburg Supercomputer Center is a part of the XCEDE system. 

The data centers then are in place and as the Doctor adds new nodes to his network, he can take advantage of the fact that IO prides itself on modular data centers that fit in shipping container like structures, and can be delivered virtually any where by truck.  A novel although opaque solution.  But one that by 2012 will mark what looks like a vertically-integrated, complete healthcare delivery system.  As he begins to do cancer genome sequencing, he will likely begin to amass considerable patient data and generate a healthy cash flow to sustain the rest of his plans.  He is generating increasing press but no one in the main stream press has looked at the totality of what he is doing.  He possess enormous power without any real oversight - something that I find worrisome.

Editor’s Note: I have tried to be transparent about this report. I have left Jen Hodson numerous phone messages.  I have never directly reached her by phone.  I sent a draft on October 13 and followed up with a message on October 21.  On October 26 she responded to an email: Hi Gordon, This is a very long report and it will take some time to look at it.  We will get back to you as soon as possible.    On October 27th she responded with a follow up question - Are you just publishing this? On your blog?   The same day I replied: A couple of clarifications.  Would you be willing to provide me with a copy of the slide deck that the Doctor used at CTIA in March of this year?

What I have on pages 48- 54  are screen shots from the video on CNBC  and they are blurred as you can see.  On October 27th she responded with a follow up question - “Are you just publishing this? On your blog?”   I responded:  It is way too long for my blog.  I publish electronicly.  See  Are you concerned?  I would HOPE that you would be pleased.  On November second, I sent Jen and Bob Peirce version 1.4 saying:  “Hii Bob and Jen, It is not yet perfect.  But the final 12 pages are substantially new right up to and including then CCMS announcements of last week.  Corrections of any errors welcome. I have reread proofed and revised the entire document.  I am sure that I have not gotten every typo but it is FAR better shape than what you last saw.

I have sent my document out for wide review - ending with version 1.10 on November 29.  Since October 27th I have had no response from  Patrick Soon Shiong’s people.  Nor from people at National Lambda Rail whom I have communicated with for years but who now appear to be effectively muzzled.

The COOK Report on Internet Protocol, (PSTN) 609 882-2572 

On Dec 4, 2011, at 8:55 PM, Gordon Cook wrote:

Bill my jaw dropped when i read this.   And i see that i had not read far enough down in your post to see what you are quoting