NSF Commits $16M To Build Cloud-Based and Data-Intensive Supercomputers

aarondubrow writes: As supercomputing becomes central to the work and progress of researchers in all fields, new kinds of computing resources and more inclusive modes of interaction are required. The National Science Foundation announced $16M in awards to support two new supercomputing acquisitions for the open science community. The systems — "Bridges" at the Pittsburgh Supercomputing Center and "Jetstream," co-located at the Indiana University Pervasive Technology Institute and The University of Texas at Austin's Texas Advanced Computing Center — respond to the needs of the scientific computing community for more high-end, large-scale computing resources while helping to create a more inclusive computing environment for science and engineering. Reader 1sockchuck adds this article about why funding for the development of supercomputers is more important than ever: America's high-performance computing (HPC) community faces funding challenges and growing competition from China and other countries. At last week's SC14 conference, leading researchers focused on outlining the societal benefits of their work, and how it touches the daily lives of Americans. "When we talk at these conferences, we tend to talk to ourselves," said Wilf Pinfold, director of research and advanced technology development at Intel Federal. "We don't do a good job communicating the importance of what we do to a broader community." Why the focus on messaging? Funding for American supercomputing has been driven by the U.S. government, which is in a transition with implications for HPC funding. As ComputerWorld notes, climate change skeptic Ted Cruz is rumored to be in line to chair a Senate committee that oversees NASA and the NSF.

Comments

Easy Fix:

[TheEyes]

Start telling Congress how Europe and China can predict hurricanes better than we can, thanks to their supercomputers. Nothing like a good "The furners are beating us!" rallying cry to squeeze money from the right (hell, the left too).

While the NSA spent $2 billion ....

[Anonymous Coward]

...on a supercomputer to spy the crap out of it's own citizens...
http://mic.com/articles/56229/the-nsa-is-building-a-new-100-000-square-foot-super-computer-to-better-spy-on-you

How pathetic can a country get?

Repuiblicans hate reality

[Required Snark]

To late. The Republican party has decided to legislate against reality already.

House Republicans pass bill forbidding scientists from advising the EPA on their own research :

H.R. 1422, which passed 229-191, would shake up the EPA’s Scientific Advisory Board, placing restrictions on those pesky scientists and creating room for experts with overt financial ties to the industries affected by EPA regulations.

The bill is being framed as a play for transparency: Rep. Michael Burgess, R-Texas, argued that the board’s current structure is problematic because it “excludes industry experts, but not officials for environmental advocacy groups.” The inclusion of industry experts, he said, would right this injustice.

But the White House, which threatened to veto the bill, said it would “negatively affect the appointment of experts and would weaken the scientific independence and integrity of the SAB.”

In what might be the most ridiculous aspect of the whole thing, the bill forbids scientific experts from participating in “advisory activities” that either directly or indirectly involve their own work. In case that wasn’t clear: experts would be forbidden from sharing their expertise in their own research — the bizarre assumption, apparently, being that having conducted peer-reviewed studies on a topic would constitute a conflict of interest. “In other words,” wrote Union of Concerned Scientists director Andrew A. Rosenberg in an editorial for RollCall, “academic scientists who know the most about a subject can’t weigh in, but experts paid by corporations who want to block regulations can.”

Re:Repuiblicans hate reality

[Anonymous Coward]

I am for separation of science and state, but not this selective kind. It is, of course, hard to attribute causes in such a complex situation, but it seems likely to me that politicization of science and science funding has played a crucial role in its gradual degradation since the 1940s. Just open any medical journal these days and all you see is people disproving strawmen. They never predict anything, they barely describe their methods, etc. I bet it is just as bad in other fields.

Re:Repuiblicans hate reality

[Anonymous Coward]

You are failing to understand one of the most basic principle of a democratic states: Separation Of Powers. No separation of state and science is required. Sciences need to become the fourth independent power inside the state.

Re:Repuiblicans hate reality

[fnj]

What are you, on crack? The Republicans in the House can pass pretty much anything they want. That's a fact of life for now. That doesn't get it through the Senate. Even after January it doesn't fly through the Senate because of the filibuster/60-vote procedure. And even if it makes it through there it sure as hell doesn't get signed by President Obama.

So really, who cares?

Re:Repuiblicans hate reality

[Curunir_wolf]

They are calling to task the "Most transparent US administration in history", which is threatening to veto this bill because "the Environmental Protection Agency should, in some case, be able to write regulations based on science and data that is not made available to the public." There's some transparency for you.

Of course propaganda from the left-wing Salon (and the only references in the article are to the even-more-left-wing ThinkProgress site) gets this bill completely wrong. Maybe try reading it?

The whole "controversy" is over some pure conflict of interest language. Exact terms: "Board members may not participate in advisory activities that involve review or application of their own work." Any of the nine board members can provide scientific data from any scientist with defense of that data; however a specific board member cannot self promote his own work. This is language found in nearly every piece of legislation that charters and administrative board. It's the same deal as requiring a judge to recuse himself it he has a personal interest in the judgment.

Anything that curtails the power of the tyrannical EPA is a good thing in my book - they have become way too powerful. If they don't have authority to regulate something, they just call up one of the many NGOs (that they provide grant money to), and tell them "Hey, you should sue us over this so we can stop it or at least tie it up in court until they run out of money or give up." If it doesn't work, at least those colluding organizations can get some settlement money out of it. That's the current SOP for the EPA, and it's undemocratic and should be illegal.

Re:Repuiblicans hate reality

[ultranova]

They are calling to task the "Most transparent US administration in history", which is threatening to veto this bill because "the Environmental Protection Agency should, in some case, be able to write regulations based on science and data that is not made available to the public." There's some transparency for you.

So, are you saying all science and data on every subject should always be public? Or that EPA should pretend not to know any that, for whatever reason, aren't?

Anything that curtails the power of the tyrannical EPA is a good thing in my book - they have become way too powerful.

Indeed, back in Good Old Days one could use water for fuel. But not to worry, fracking is already working on getting them back.

But hey, maybe you fancy living in Mordor.

Re:Repuiblicans hate reality

[Curunir_wolf]

So, are you saying all science and data on every subject should always be public?

Nice straw man you've got there. Obviously, I never said or intimated anything of the sort. If some idea is being used to create public policy, enforced by armed bureaucrats, then, yes, absolutely, the science needs to be public and available.

Indeed, back in Good Old Days one could use water for fuel

Another nice straw man. Oh, right, since we had worse pollution 70 years ago, every little tyrannical behavior of the EPA should be allowed without question.

But hey, maybe you fancy living in Mordor.

... and you fancy living in North Korea. nyah.

Re:Repuiblicans hate reality

[micahraleigh]

Harry Reid wants you to know only a straight majority matters now.

... As he hands over the majority leader status.

Re:Repuiblicans hate reality

[micahraleigh]

As Wittgenstein (a very liberal guy) noted: reality is a personal construct.

The happy man lives in a different universe than the unhappy man. Etc.

Re:Repuiblicans hate reality

[tnk1]

"Sciences" should not be an independent anything in the state.

You want to keep science as far from government as you can. Indeed, this only shows that its too close already.

What kind of fucking shit is this?

[Taco Cowboy]

If TFA wants to talk about supercomputer then it should stick to supercomputers. What is the point of sticking in the following:

... climate change skeptic Ted Cruz is rumored to be in line to chair a Senate committee that oversees NASA and the NSF

1, TFA has failed to substantiate its claim that Ted Cruz is a climate change skeptic
 
2. Even if Ted Cruz is a climate change skeptic it still has NOTHING to do with the funding of supercomputers
 
3. Supercomputer can be used for many things, not only for climate pattern modelling
 
4. TFA also failed to prove that Ted Cruz has refused to fund the purchase of supercomputer
 
5. Even if Ted Cruz is such a villian, that the guy would end up destroying NSF and whatnots, why didn't the American government, led by the Obama administration and its Democratic party congress critins, before Ted Cruz arrives at the scene, fund the purchase of supercomputers?
 
Is TFA going to be an article about the lack of supercomputer or is TFA going to be a political character assassination propaganda against Ted Cruz?
 
I have no relationship whatsoever with Ted Cruz. I just can't stand this kind of below-the-belt sneak attack disguising as a Slashdot article

Re:What kind of fucking shit is this?

[Anonymous Coward]

The point is with Cruz in place the situation is only going to get worse. It's fair game.

Re:What kind of fucking shit is this?

[Anonymous Coward]

2. Even if Ted Cruz is a climate change skeptic it still has NOTHING to do with the funding of supercomputers 3. Supercomputer can be used for many things, not only for climate pattern modelling

While I can't speak about Ted Cruz, when talking to a Senator touring the research done at the university I work at, they cut me off when describing the uses of our HPC equipment to ask if it was used for climate modeling too. After confirming that it was used for that, there was some quip about how weathermen are still wrong despite all of the new computers, and then moved on ahead of schedule. Don't underestimate the ability of politicians to focus on a single part of a multifaceted project, or at least appear to because they don't care about any facet.

5. Even if Ted Cruz is such a villian, that the guy would end up destroying NSF and whatnots, why didn't the American government, led by the Obama administration and its Democratic party congress critins, before Ted Cruz arrives at the scene, fund the purchase of supercomputers?

HPC projects have received quite a lot of funding over the past years. That doesn't stop new projects from coming up or the requirement of on going funding.

That said, the political reference seems gratuitous for me too.

Re: New Study Confirms It

[Anonymous Coward]

Apparently there are those who ride with serial killers. Only you never about them. Anymore.

We need more 3 letter organisations.

[Anonymous Coward]

Why don't we, (WDW), just speak in, (JSI), three fucking letters? (TFL)?
That way, NSF can just NSA.

Re:New Study Confirms It

[Anonymous Coward]

BRING BACK THE TAGLINE: NEWS FOR NERDS, STUFF THAT MATTERS.

Dice is a serial killer (of Slashdot culture). Also, FUCK BETA.

Cloud-Based, Data Intensive,Super Computer?

[wisnoskij]

So what does that mean? They are going to spent $16M on a contract with Amazon to have them host some data in the cloud?

Re:Cloud-Based, Data Intensive,Super Computer?

[Nite_Hawk]

Amazon is a possibility for some research (and there are PIs who haven gone that route). There are a couple of problems:

1) If you use EC2 24/7 and need a ton of data storage and fast data transfer capabilities it's no longer that cheap.

2) Sending potentially sensitive data off to amazon servers isn't a great idea. Even if you have data that is supposed to be de-identified, there are PIs who will intentionally or unintentionally screw up and put sensitive data on your cluster. It's one thing if this is inside an academic lab. It's another thing entirely if it's beamed over the internet to uncontrolled machines.

3) The amount of data being collected these days is mind-bogglingly huge. Even a couple of years ago when I was more directly involved in HPC, data sets for things like genomics data where gigantic. They could collect several TB of data per day and it was rapidly increasing. Transferring all of that off to amazon takes a lot of bandwidth and time. Keeping the cluster closer to the data collectors can be a win.

Re:Cloud-Based, Data Intensive,Super Computer?

[tnk1]

I'd agree that there are certain applications that AWS would be not so good for.

If you can guarantee nearly full utilization of your equipment and a certain amount of bandwidth, you're probably best with buying your own stuff. The price tag will be large up front, but you'll save money over the long term.

That said, if the problem is ramping up and down your needs, you might have a good case for at least adding Cloud services to your existing mix. It is probably a bitch to get new equipment for the government, as well as getting it approved and set up. If you can scale out in the cloud, you have the ability to codify your infrastructure and recreate it at will, when required. That means not only almost instant set-up of networking and instances, but also near instant set up of software and security features (if needed). And when you don't need it, blow it away, and the infrastructure code will be there the next time you need it.

Initial data loads into EBS or S3 can be done via drop shipping hard drives to AWS, so bandwidth for initial loads may not be as bad as it would appear.

Cheap Mutha

[JimSadler]

Instead of a paltry 19 million why not toss them a billion bucks to build a computer that will rock the world. Apparently some nations now engage us in forms of cyber warfare daily. So lets not get slightly ahead. Let's get Star Wars types of ahead and build a computer that can easily handle any threats thrown at it. And such a machine just might get rid of problems like cancer, poverty and other plagues that mankind has always endured.

Re:Cheap Mutha

[tnk1]

Cancer is a a tough one because it is not actually one disease, but many causes for abnormal cell division and tumor growth.

Most of the other ones are actually not things you need computers to fix. Except for temporary situations caused by disasters, the only real reason for plagues and poverty is human nature.

We could feed the entire planet with the food we're growing now. Everyone. Instead we burn it or turn grains into ridiculous forms of fuel or drinks because we don't care to share it, and what is more, many times when we do want to share, the powers that be on the other side don't want to let us give away that food.

Technology has already advanced us to this point where we could legitimately feed and clothe people, but it is never going to do that until we allow ourselves, and they allow us, to take care of one another. Unfortunately, that means we'd need to get over certain things like antagonistic relationships and greed to get there. And not just corporate greed. All forms of greed and distrust. It may seem strange, but there are people who will starve today because they don't want to accept help from someone they don't trust. The choice is not always that obvious, but decisions like that are made all the time in the name of nationalism or pride or (again) greed.

the skeptic is ... who?

[micahraleigh]

Considering that only 42% or so of Americans believe global warming is real, I'm not sure why Cruz is getting labeled a "skeptic".

Or do the opinions of ordinary people count for nothing?

Re:the skeptic is ... who?

[ndykman]

When compared to the broad consensus of science, yes. Belief doesn't enter into it, the research is done. Global warming is an established fact. And not just by one paper, but by repeated, peer reviewed research. Even early skeptics in climate modeling have come to the same conclusions.

I hesitate to call him or others skeptical, as it suggests there is really any room for doubt. There really isn't. The core findings about global warming are established. Covering our ears and shouting "it's not true" won't change a thing.

Re:the skeptic is ... who?

[Anonymous Coward]

Or do the opinions of ordinary people count for nothing?

Considering the number of people who think antibiotics kill viral infections or that the sun orbits the Earth...

Re:the skeptic is ... who?

[micahraleigh]

I don't think you realize how much your posture is damaging your cause.

When people see "the experts" insisting, pushing, fighting, and demonizing that they understand what ordinary people don't ... people stop listening.

The articles are peer reviewed so dissenting voices are kept out. See climategate.

The hockey stick graph has been thoroughly discredited.

Global temperatures have been dropping more than they've been increasing.