×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

NWS Announces Big Computer Upgrade

timothy posted about a year ago | from the in-these-troubled-economic-times dept.

Earth 161

riverat1 writes "After being embarrassed when the Europeans did a better job forecasting Sandy than the National Weather Service Congress allocated $25 million ($23.7 after sequestration) in the Sandy relief bill for upgrades to forecasting and supercomputer resources. The NWS announced that their main forecasting computer will be upgraded from the current 213 TeraFlops to 2,600 TFlops by fiscal year 2015, over a twelve-fold increase. The upgrade is expected to increase the horizontal grid scale by a factor of 3 allowing more precise forecasting of local features of weather. The some of the allocated funds will also be used to hire some contract scientists to improve the forecast model physics and enhance the collection and assimilation of data."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

161 comments

Go suck Obamas dick you progressive shit birds (0, Troll)

Anonymous Coward | about a year ago | (#43772449)

Obama fundamentally transforms America (a brief primer):

There was a time when the IRS acted against Al Capone. Now, under Obama, the IRS acts like Al Capone.

There was a time when the FBI investigated crime syndicates. Now, under Obama, the FBI investigates for a crime syndicate.

There was a time when the ATF stopped Gun Running Gangs. Now, under Obama, the ATF is a Gun Running Gang.

There was a time when the TSA protected women at airports. Now, under Obama, the TSA molests women at airports.

There was a time when people in India dreamed of coming to America to get Health Care. Soon now, under Obama, people in America will dream of going to India to get Health Care.

Re:Go suck Obamas dick you progressive shit birds (0)

Anonymous Coward | about a year ago | (#43772465)

Good troll except for the TSA line. The TSA has never protected anyone in greater proportion to the harm they cause.

Re:Go suck Obamas dick you progressive shit birds (0, Offtopic)

EmagGeek (574360) | about a year ago | (#43772551)

I dunno, I thought the IRS line was pretty fucking accurate.

Use the old one to mine bitcoin (4, Funny)

Anonymous Coward | about a year ago | (#43772481)

maybe it would pay for the upgrade.

Precise garbage (2, Insightful)

Anonymous Coward | about a year ago | (#43772495)

Well unless we're in a butterfly wing effect situation, you'll generate 3 times the amount of garbage.

I think this is more a subsidy to the troubled supercomputing market disguised as a technical improvement.

Re:Precise garbage (0)

Anonymous Coward | about a year ago | (#43772531)

Why do you say that?

The Europeans did better, and they do have more computing power. It doesn't prove the greater computing power is why they did better, but it's not unreasonable.

No, Europe had 50 TFLOPS, 1/5th the USA (5, Interesting)

Anonymous Coward | about a year ago | (#43772631)

http://www.ecmwf.int/services/computing/overview/supercomputer_history.html

Europe: 70 TFLOPS by and upgrade to be finished by early 2013 (Sandy was in Oct 2012), which they say will make it about 3 times the power of the computer it replaces. i.e. 23 TFLOPS, they did a part upgrade during Sandy, to about 50 TFLOPS

USA: 213 TFLOPS, to be upgrade to 2,600 TFlops

So no, the Europeans did the prediction with 10%-20% of the supercomputing power, 2% of the proposed supercomputing power. This is just a subsidy to the Supercomputer industry (and indirectly USA chip makers), at a time when the PC market is tanking. It has nothing to do with the garbage the US produced, they just used a bad model.

"Replacement of the second cluster will be completed in early 2013. Each cluster has 768 POWER7-775 servers connected by the IBM Host Fabric Interface (HFI) interconnect."

"For the first time the processor clock frequency actually decreased, going from 4.7GHz to 3.83GHz, despite this each processor core has a theoretical peak performance 60% greater than that of the POWER6. For ECMWF's applications the system is about three times as powerful as the system it replaced.The first operational forecasts using this system were produced on 24 October 2012."

Re:No, Europe had 50 TFLOPS, 1/5th the USA (3, Insightful)

operagost (62405) | about a year ago | (#43773183)

This is similar to how politicians and teachers' unions insist that the way to produce better results in our public schools is to throw more money at them. Meanwhile, the performance of European schools (and even American private schools) do better with less.

Re:No, Europe had 50 TFLOPS, 1/5th the USA (1, Insightful)

Anonymous Coward | about a year ago | (#43773575)

The assertion that private schools are better is a dangerously miguided one for a few reasons:

1) private schools can and do choose the students they want. So they either throw out or do not enroll the students that take the most time and effort. This includes for things like discipline problems, learning disabilities, and phisical disabilities. So they avoid all of the expensive students. And those students are enormously expensive to take care of.

2) We have let the "special needs" system start to gobble up enormous resouces, primariyly in legal and administrative (not teaching) costs. My mother was a special ed. teacher and I was constantly amazed at all of the large mettings she described covering the IEP (Individual Education Plans) where most of the people at the table were not teachers, and there were often multiple lawers at each meeting for the school district. At a couple of hundred dollers per hour each, those are going to add up. If we were to start making reasonable limitations on what parents could sue the school districts for that would cut an enormous chunk out of the school budgets.

3) Private schools are not required to take the same tests as public schools. So it is hard to have any real data about them doing better. Every time I have heard someone assert than and have been in the position to ask, it winds up to be an unbased assertion.

4) In addition to the private school selecting their students, the students also tend to be self-selecting for those with both well-off parents as well as involved parents. Leaving asside arguments about inherited raw ability, those parents have the skills needed for success as well as the motivation. They are likely to teach those to their children regardless of where those children go to school. So they are pooling students who are likely to perform highly, as well as parents who are highly involved with the education system away from public schools.

5) Many private schools are religious ones. So a portion of their teachers and administrative staff have already given vows of poverty. Funny how that makes it easier to have a smaller budget.

Public schools are the only thing this country currently has that promotes equality of opportunity, and every generation there is a big push to try and kill them and replace them with ways that the rich can secure their children's future at the expense of the poor children's.

Re:No, Europe had 50 TFLOPS, 1/5th the USA (5, Insightful)

Anonymous Coward | about a year ago | (#43773991)

So much ignorance, I'm not sure where to start. First, I work in a public school. I've worked in private schools, my mother runs a (non-religious) private school. My wife has taught in other private and public schools. My daughter has an IEP.

Teacher's unions generally do not want more money thrown at the schools, depending on the state. There will be significant differences in the political games in public education depending on the state. In most "at-will employment" states, the teacher's union is mostly there for show. The district mostly controls the teacher's union. In states like this, the district administration pushes for more money to be thrown at "schools." This is because they control how the money is actually spent, and as a result, most of the money doesn't get to the school level.

There is a huge misdirection that most of the general public has fallen for in public education. The perception among the general public is that the "schools" are at fault. In reality, the district administration controls and dictates everything. A school principal has much less authority and autonomy than most people realize. This works out great for the district administration, because the school staff regularly become the scapegoat for failed district policies. In many states, counties, districts, cities, a school can do very little other than what district administration tells them to do. In effect, a school has all the accountability with none of the authority. Meanwhile, the district administration continues to make decisions in a vacuum while collecting paychecks that would make a seasoned IT Director blush.

I've been in countless IEP meetings, both as a parent and as a school administrator. Most IEP meetings are educators and parents. 4-5 school staff, 1-2 parents. The school staff are usually the various specialists (speech, OT, learning specialist, etc) and general ed teacher. I will usually be involved if there's some behavior concerns related to the IEP. I have never had a lawyer in an IEP meeting, other than a child advocate when there's social services involvement with a student. If a district needs lawyers at every IEP meeting, they're doing something very wrong. That would suggest the district is the problem, not the system itself.

One last note, socio-economic status has a larger impact on student success than most people want to admit. Districts don't want to talk about that because then they might lose the federal programs thanks to No Child Left Behind.

tl;dr It's not the individual public schools that are a problem, it's district administrations and school boards that have created huge bureaucratic structures and keep huge portions of money at the district level. This is really why private and charter schools generally do better with less money. They don't have the huge bureaucracy sucking up the money.

Re:No, Europe had 50 TFLOPS, 1/5th the USA (1)

mozumder (178398) | about a year ago | (#43774005)

This is similar to how politicians and teachers' unions insist that the way to produce better results in our public schools is to throw more money at them

I guess some people think you an get better results for free.

Re:No, Europe had 50 TFLOPS, 1/5th the USA (0)

Anonymous Coward | about a year ago | (#43773289)

Europe: 70 TFLOPS ... USA: 213 TFLOPS, to be upgrade to 2,600 TFlops

The link you provided says the latest European system has 70 Tflops sustained and 1.5Pflops peak. That seems like an awfully big ratio. Is this a typo or is that the way current supercomputers are? More to the point, the WaPo article doesn't say if the 213/2600 Tflops of the old/new American systems is peak or sustained.

Re:No, Europe had 50 TFLOPS, 1/5th the USA (1)

Anonymous Coward | about a year ago | (#43773315)

at http://www.ecmwf.int/services/computing/overview/ibm_cluster_phase2.html we can see that ECMWF lists 70 TFLOPS as the SUSTAINED performance of their system, wheras the US numbers (213 TFLOPS and 2.6 PFLOPS) are peak. Big difference.

Not true, both are sustained (0)

Anonymous Coward | about a year ago | (#43773667)

Not true,

The current USA system is 70 TFLOPS, this summers upgrade takes it to 213 TFLOPS.
Europe had 23 TFLOPS sustained at the beginning of Sandy, their upgrade early in the year takes them to 70 TFLOPS to catch USA.

USA has announced an upgrade from 213 to 2600 by 2015...

http://cliffmass.blogspot.com/2013/05/a-new-chapter-for-us-numerical-weather.html

"Using 24 millions dollars from the Superstorm Sandy Supplemental budget, the National Weather Service will be acquired two computers with a capacity 37 times greater than it uses today. We are talking about a transition from 70 teraflops right now (and 213 teraflops this summer) to 2600 teraflops in 2015. "

Re:No, Europe had 50 TFLOPS, 1/5th the USA (3, Informative)

ebno-10db (1459097) | about a year ago | (#43773585)

You're confusing peak and sustained performance. According to the link you provided the latest European system has 70 Tflops sustained performance, but 1.5 Pflops peak. According to this article [theregister.co.uk] the ratings given for the American systems are peak, so the European system is much more powerful than the current US system.

RMax is 178TFLOPS (0)

Anonymous Coward | about a year ago | (#43773999)

http://www.top500.org/site/48440

Well no, their numbers (referred to in the Register article) show they have a sustained max of 178TFLOPS, peak of 213.

Europe had an older system at the start of Sandy, 1/3rd of 70TFLOP sustained = 23 TFLOPS.

So USA had 8 times the processing power.

Re:Precise garbage (2)

Shavano (2541114) | about a year ago | (#43772643)

Did they do better because of
  • more computing power?
  • more accurate models?
  • dumb luck?

Re:Precise garbage (1, Interesting)

Anonymous Coward | about a year ago | (#43772691)

The Europeans rightfully use Fortran for the numerical simulations, while the US hipster-doofus coderz use C and tons of flying pointers everywhere (essentially just sophisticated GOTOs). This creates code that is far, far less efficient. I wouldn't be surprised if much of their C codebase has been refactored with the use of automated tools several times.

OP is making a statement for one of those (0)

Anonymous Coward | about a year ago | (#43773109)

OP is making a statement for one of those being wrong.

So you'd better ask the OP how they know that it wasn't because of more computing power. GP doesn't say that it must be that, but the OP says it cannot be. Yet you demand that the non-positive claim must be proven???

3DVAR vs. 4DVAR vs. better models (2)

tlambert (566799) | about a year ago | (#43773915)

This upgrade in computing power is to move the US from 3DVAR to 4DVAR, however, it does nothing to improve the US weather models. This is interesting, in that 4DVAR can give worse results than 3DVAR, while using additional compute power. There was a nice paper written in this:

http://onlinelibrary.wiley.com/store/10.1256/qj.05.85/asset/200513161304_ftp.pdf [wiley.com]

"The first proviso is that on other measures of analysis quality the conclusions are less clear cut. For instance, Fig. 5 shows that using evolved covariances gives worse fits for synoptic 4D-Var than 3D-Var with FGAT. Despite the 4D-Var schemes giving better forecasts overall, their analyses are not consistently better. An explanation may be that the evolved covariances used in 4D-Var (see appendix) increase the background error and hence the impact of observations in modes growing in the time-window, and that this improves the analysis and hence the forecast if these modes continue to grow. This effect was discussed by The paut et al. (1996) who showed that the evolved covariances implicit in 4D-Var are very similar to the singular vectors representing the fastest growing modes. This effect is strongest for longer evolutions; most of The paut et al.’s results are for 24 hours rather than the 3 hours in this paper. For a 6-hour window they see some decaying modes, which they attribute to discrepancies between the 3D- Var covariances and the PF dynamics which is used to evolve them. There are also probably biases in the simple PF model, which may distort the evolved covariances. For these decaying or distorted structures 3D-Var has more freedom to fit the observations, so 4D-Var analyses can be worse."

Yes, that was precisely garbage. (0)

Anonymous Coward | about a year ago | (#43773035)

How much of Sandy was due to the wingbeats of the Monarch Butterfly that I saw last summer?

And if the weather/climate is so sensitive that a butterfly's wings REALLY CAN cause hurricanes, then how the hell do you straighten that with the demand that humans cannot cause any significant change because CO2 is a trace gas??? It's a fucking huge mass of CO2, even compared to the total weight of all the butterflies that could possibly exist. INCLUDING the body, which doesn't flap...

Re:Precise garbage (0)

Anonymous Coward | about a year ago | (#43773701)

Well... except for the fact that hurricane prediction ability has been quantitatively improving for the last several decades... I guess the secret cabal of weather people must have taken up butterfly training instead of actually making objective progress in modeling...

horray! (0)

Anonymous Coward | about a year ago | (#43772497)

Upgrading to Pentium 4s at only $500,000 each! What a bargain!

with the 10x boost in CPU (0)

Anonymous Coward | about a year ago | (#43772521)

The boxen will be able to synthesize video of a weatherman making Carson-style jokes about how badly they missed last night's forecast.

so.. (5, Funny)

msauve (701917) | about a year ago | (#43772537)

Why not just pay attention to the European forecasts, which would cost nothing?

Re:so.. (4, Informative)

BRock97 (17460) | about a year ago | (#43772623)

Why not just pay attention to the European forecasts, which would cost nothing?

Actually, the NWS pays a great deal of money to see the ECMWF (the European model of choice) and are required to encrypt it before it is sent out to the various forecast offices over their NOAAPort system.

Re:so.. (0)

Anonymous Coward | about a year ago | (#43772633)

Because crunching data like this is intensive and Europe probably wants to focus on itself, seeing as it wanted to build the computers capable of doing this in the first place.

I imagine that, on seeing a big hurricane coming, or maybe just because it was hurricane season itself, they devireted extra resources to predict Sandy.

They probably charge just enough to insentivise other continents to build their own similar systems.

not that I know anything about the details but I would imagine it's probably along those lines

Re:so.. (0)

Anonymous Coward | about a year ago | (#43772635)

because "we didn't earn it", and "we did not build it", and "that's socialism", and "France"

Re:so.. (3, Insightful)

Teun (17872) | about a year ago | (#43772969)

Because, as others already explained, the NWS is already using the EU generated data and vice versa.

Much more important is for something this important you'd like to have more than one model, if only to check against.

23.7 instead of 25M (1, Funny)

Anonymous Coward | about a year ago | (#43772539)

Looks like they'll have to shut down most essential weather services. The People have to be taught a lesson about telling government to spend less money.

The reason is pretty lame (2)

houghi (78078) | about a year ago | (#43772563)

Because Europe was better? Why not:because we want to increase our quality regardless of what others are doing. Think about it: if the Europeans would not have been better, what you had would have been good enough. Or "We could predict the storms better and save potential lives, but who really cares? We are already the best in doing it. USA! USA! USA!"

Re:The reason is pretty lame (-1)

Anonymous Coward | about a year ago | (#43772647)

Computational fluid dynamics has many other applications. At current prices, 2600TFlops costs around $12 million (62 of these [compu-america.com]).

WTF? They're wasting my tax money.

Re:The reason is pretty lame (2)

Cenan (1892902) | about a year ago | (#43772759)

"Hey, a storm is coming"
"Ok, so what do we do?"
"Dunno, i spent the money being able to tell you that a storm is coming"

Re:The reason is pretty lame (1)

amck (34780) | about a year ago | (#43772967)

"Hey, a storm is coming"
"Ok, so what do we do?"
"Dunno, i spent the money being able to tell you that a storm is coming"

"So what do we do?"

Evacuate. Seal buildings. Prep all emergency services (make sure none of the fire engines, ambulances, etc. are in the shop).
Cancel leave.

This all costs money: the further ahead and the more precise you can forecast the storm track, the less it costs.
And yes, the NWS will have had to provide good evidence they can save that money in order to justify the upgrade.

Re:The reason is pretty lame (1)

Medievalist (16032) | about a year ago | (#43774305)

I swear, the way things are going, I expect people to start living on platforms suspended over active volcanoes and demanding taxpayer dollars for their air conditioning costs.

If you live somewhere that nature has decided is no longer going to be habitable by humans, get out or go down with the damn ship but either way do not expect anyone to help you rebuild in the same place. The most the taxpayer should be on the hook for is helping you relocate, which is generous enough. Evacuate permanently, or not at all.

Re:The reason is pretty lame (0)

Anonymous Coward | about a year ago | (#43773003)

Unfortunately compu-america.com only has 10 of those in stock at the moment. We'll have to find a supplier willing to provide some big kickbacks to get the rest at a much higher cost.

Ever seen the movie Seabiscuit? (0)

Anonymous Coward | about a year ago | (#43773077)

America is the same way. The only tried and true motivational method that works is giving them a perceived worthy competitor. If there is not one available then America will do nothing at best. Just look at NASA. Made great strides when there was somebody better. Once their opponent stagnated NASA started to self destruct.

The key in getting them to do anything for the good of humanity is to show them in a meaningful way that someone else has done it better. None of this "EU reports they are better than USA at *foo*". All that those things do is stir up politicking. The example needs to be concrete, something you can point to. Then and only then will you see the massively competitive attitude combined with economic resources being applied to something worthwhile.

You are all missing the REAL reason (0)

Spritzer (950539) | about a year ago | (#43773995)

The NWS doesn't need a faster supercomputer. The current one can pump out bad results based on a flawed set of algorithms at a perfectly useable rate. What the current computer can't do is act as an East Coast regional processing point for THIS [slashdot.org].

It's a good day to be a physical oceanographer (1)

superflippy (442879) | about a year ago | (#43772581)

Not that the 5 people in that field have trouble getting jobs anyway. But if you like math and the ocean, it's a good field to go into.

Things you find out after the fact ... (2)

fygment (444210) | about a year ago | (#43772617)

... the Europeans did a better job forecasting Hurricane Sandy. Oh. Didn't know that. But hey when they make a movie of it, I'm sure they will present as fact that the American system was the most awesome thing and NWS was right on the money with typical awesome American ingenuity .... sorry, 'Argo' flashback.

Re:Things you find out after the fact ... (1)

tysonedwards (969693) | about a year ago | (#43774039)

To the same end, on historical analysis there has been one model that has held true more often than it has not.

The weather tomorrow will be exactly the same as today (+/- 1%)

Sure it is a meteorological and mathematical joke, you can't argue with the results.

The most needed upgrade (2)

alantus (882150) | about a year ago | (#43772625)

Hopefully this new system will be able to calculate in Celsius instead of Fahrenheit...

Re:The most needed upgrade (0)

Anonymous Coward | about a year ago | (#43772687)

Though most of the world uses the Celsius scale, the Fahrenheit scale may be better suited to meteorology. For one thing, it is more precise and less coarse simply because each degree represents a smaller interval.

More importantly, the range in temperature from 0 to 100 degrees Fahrenheit almost perfectly demarcates the extremes found in the climates of the United States and Europe; it seldom gets any hotter or colder. The convenience of a perfect 100 degree interval encompassing the temperatures in which most of us live seems a pity to lose. (The same range on the Celsius scale is a clumsier -18 to +38 degrees.)

Re:The most needed upgrade (4, Insightful)

Cenan (1892902) | about a year ago | (#43772809)

Though most of the world uses the Celsius scale, the Fahrenheit scale may be better suited to meteorology. For one thing, it is more precise and less coarse simply because each degree represents a smaller interval.

Bullshit. There is no precision to be had from choosing a unit, the precision comes from not being an idiot and doing all your calculations in straight integers.

More importantly, the range in temperature from 0 to 100 degrees Fahrenheit almost perfectly demarcates the extremes found in the climates of the United States and Europe; it seldom gets any hotter or colder. The convenience of a perfect 100 degree interval encompassing the temperatures in which most of us live seems a pity to lose. (The same range on the Celsius scale is a clumsier -18 to +38 degrees.)

More bullshit. The argument is based on how you feel towards a given range, but nobody is going to do those calculations by hand. You could just as easily have a range of 0-1 and have the exact same precision as before, just more numbers after the decimal point.

And predicting the weather is not about predicting the normal as much as predicting the extremes, which would lie outside your "perfect range".

Weather reports are in units of degrees. (1)

Anonymous Coward | about a year ago | (#43773285)

You're talking as a meteorologist or computer programmer.

He's talking as 90% of the US public.

I think you need to be a little less a pompous ass and a little more understanding.

Re:Weather reports are in units of degrees. (0)

Anonymous Coward | about a year ago | (#43773499)

Both scales have certain advantages. All of those advantages are so trivial as to be negligible. For talking about everyday weather, I find Fahrenheit slightly superior.

Re:Weather reports are in units of degrees. (1)

drinkypoo (153816) | about a year ago | (#43773555)

The everday public benefits from not having to think about fractional degrees, which can't even be felt on the Fahrenheit scale.

Really its only benefits are that the numbers are convenient for us to think about in everyday use while not doing a lot of mathematics with them. But those are benefits.

Meanwhile, why not switch over to K?

Re:Weather reports are in units of degrees. (0)

Anonymous Coward | about a year ago | (#43773675)

Think about a truly decimal "0 = freezing, 10 = boiling" scale. If someone tried to impose that scale on us and started saying things like, "What's the problem? It's easy. Zero is freezing, one is not, two is pleasant, three is hot. If you need more precision, just use decimal points," I doubt they'd be taken very seriously.

Re:The most needed upgrade (2)

sidyan (110067) | about a year ago | (#43772921)

AC is in for a wonderful discovery: Real numbers [wikipedia.org]

Re:The most needed upgrade (0)

Anonymous Coward | about a year ago | (#43773319)

Speaking as an EE, I see no reason to prefer "real" numbers to imaginary ones.

Re:The most needed upgrade (1)

Teun (17872) | about a year ago | (#43772983)

Haha, never heard of decimals because you still use fractions?

Besides, when it comes to science it wouldn't surprise me they use an absolute scale like Kelvin.

Re:The most needed upgrade (0)

Anonymous Coward | about a year ago | (#43773761)

The Rankine scale also starts at absolute zero, a la Kelvin, but uses Fahrenheit-sized degrees. Kelvin and Rankine are equal at absolute zero.

Re:The most needed upgrade (1)

Runaway1956 (1322357) | about a year ago | (#43773479)

You can't be frigging serious. Really, you can't be. Precision? Do you even understand what the word means? And, "clumsy numbers"? You're auditioning for Seinfeld or something, right?

Re:The most needed upgrade (1)

Sockatume (732728) | about a year ago | (#43773481)

On the other hand, zero celcius is the boundary line between dealing with frozen water (as ice or snow) and dealing with liquid water (as flooding or rain). That's incredibly convenient when travelling. I don't think that the nuanced subtlety implied by indicating that it's going to be 95F instead of 94F tomorrow is really worth the tradeoff, or for that matter reflected in the precision of the model itself.

Arbitrary scales (1)

sjbe (173966) | about a year ago | (#43773619)

On the other hand, zero celcius is the boundary line between dealing with frozen water (as ice or snow) and dealing with liquid water (as flooding or rain). That's incredibly convenient when travelling.

Not that I'm in favor of fahrenheit (I'm not) but water freezing at 32F isn't any more difficult to deal with than 0C. Both are arbitrarily chosen chosen scales. Celsius has the nice round numbers but from a practical day to day usage standpoint that matters not at all. I know that water freezes below 32F and that doesn't take up any more room in my brain than 0C. The only real problem is that I have to remember two scales instead of one. Since Celsius is the more widely used scale, I wish we would switch to that one.

Re:The most needed upgrade (2)

Sockatume (732728) | about a year ago | (#43773515)

Don't think of -18 celsius to +38 celcius, which really is clumsy; think of -20 to +40 celcius.

-20 to -10 is a deep freeze

-10 to 0 is freezing

0 to 10 is cold

10 to 20 is warm

20 to 30 is hot

30 to 40 is danger

I would start looking at the algorithms (5, Insightful)

Chrisq (894406) | about a year ago | (#43772641)

It appears that the computers that Europe was using for the "better forecast" were not as powerful [ecmwf.int] as the [metoffice.gov.uk] old system being replaced. Upgrading because Europe's forecast better would be like taking a slow route to a holiday destination then buying a Porsche because your neighbours got there sooner when all you need is a new roadmap.

Re:I would start looking at the algorithms (3, Interesting)

Chrisq (894406) | about a year ago | (#43772681)

Also, though I would like to believe that Europeans have superior algorithms, realistically the hurricane prediction could be a "one off". We know that modeling weather can gibe widely different results based on small variations of starting conditions, assumptions, etc. Unless there is evidence that European forecasts are consistently better it could just be luck. With the known chaotic nature of storm systems it wouldn't surprise me if the "butterfly effect" of the rounding errors when converting from C to F would be enough to displace a storm by hundreds of miles!

They run multiple scenarios (1)

Anonymous Coward | about a year ago | (#43772755)

To detect butterfly effects they bracket the scenarios with a small delta and see if it swings off chaotically. If that happens, then they know they can't make a realistic prediction, because the sensors they have don't permit it. Adding more computing power doesn't fix anything then.

No, it was simply a little better algorithm run on a computer a tiny fraction of the processing power of the *current* US supercomputer. I notice there's a lot of government money going into supercomputers as the PC market dries up. I look at the use case and they're a joke. Similar to this one.

I think its just a case of pumping money into the US semiconductor industry and IBM at a time when they're struggling.

Re:I would start looking at the algorithms (4, Informative)

Shinobi (19308) | about a year ago | (#43772803)

It's not just once. Several hurricanes and other severe weather systems have been most accurately predicted by the European model. In fact, if you read some of the links in the article, you'll see references to that.

Re:I would start looking at the algorithms (3, Informative)

Overunderrated (1518503) | about a year ago | (#43773203)

With the known chaotic nature of storm systems it wouldn't surprise me if the "butterfly effect" of the rounding errors when converting from C to F would be enough to displace a storm by hundreds of miles!

Absolutely not the case. First, all non-trivial computational fluid dynamics codes (e.g. those used for weather prediction) use non-dimensionalized values in the governing equations. You're not solving with C vs F (you'd never use either anyway, but absolute kelvin vs rankine), or meters vs feet, but non-dimensional numbers which are only converted back to dimensional ones after all the heavy computation is done.

Secondly, even if one were to use dimensional values in solving the equations, the round off errors for converting between C and F are many, many orders of magnitude smaller than the errors you get in the discretization of the original continuous system of equations.

Lots of comments here regarding metric vs. imperial units; I assure you that accuracy discrepancies between the European and American predictions have absolutely nothing whatsoever to do with any choice of unit system.

Source: I'm a CFD researcher =)

Re:I would start looking at the algorithms (2)

ColdWetDog (752185) | about a year ago | (#43773373)

Read the "Scientific Weather Discussion" in Weather Underground forecasts. More often than not, they find that the ECMWF forecast fits the data better. Been that way for years.

Re:I would start looking at the algorithms (0)

Anonymous Coward | about a year ago | (#43772815)

Yeah, I can't see them putting money into pure climate and weather modelling research. A sizable fraction of the Congress thinks that everybody in that field is a con artist and all their results are fake. It'd be about as easy to push through as research into the effects of gun control.

A fast new computer has a nice level of separation from the troublesome science. Maybe it'll end up crunching some sophisticated algorithms from people who are allegedly members of the biggest conspiracy ever, but at least Congress won't actually have to take on the cognative dissonance directly.

Re:I would start looking at the algorithms (1)

slimdave (710334) | about a year ago | (#43772871)

US weather forecasters tend to agree with Congress. http://uk.reuters.com/article/2012/02/02/us-climate-weather-idUSTRE81120K20120202 [reuters.com]

Re:I would start looking at the algorithms (1)

ebno-10db (1459097) | about a year ago | (#43773371)

No, that article is drumming up controversy that doesn't really exist. From TFA:

Only 19 percent of U.S. meteorologists saw human influences as the sole driver of climate change in a 2011 survey.

I'm surprised it isn't 0%. The vast majority of climate scientists don't believe human influence is the sole cause either. Considering how much the climate has changed w/o human intervention, it's ridiculous. One of the difficulties of convincing people of AGW is that it's superimposed on a natural warming trend (emergence from the little ice age and beyond).

We've been out of that recovery for centuries. (0)

Anonymous Coward | about a year ago | (#43774023)

Here's a clue for you. Made up temperatures, though they're "approximately" correct, to show you the deal.

Before that ice age, global temperature averages were 13C. During the middle of that ice age, global temperature averages were 11C. We recovered to 13C in the 1800's.

Meanwhile, on the 10,000 year scale, we are *going in to* an ice age if you discount human efforts, so the trend should be DOWN ~0.2C per century, rather than UP 2.0C per century.

Re:I would start looking at the algorithms (0)

Anonymous Coward | about a year ago | (#43773867)

I would RTFA:

1. The U.S. has inadequate computer power available for numerical weather prediction. The ECMWF is running models with substantially higher resolution than ours because they have more resources available for NWP. This is simply ridiculous--the U.S. can afford the processors and disk space it would take. We are talking about millions or tens of millions of dollars at most to have the hardware we need. A part of the problem has been NWS procurement, that is not forward-leaning, using heavy metal IBM machines at very high costs.

Good (2)

macwhizkid (864124) | about a year ago | (#43772657)

It's kind of astonishing how little we (by which I mean the U.S.A.) spend on weather forecasting relative to the economic effects. The economic costs of weather are in the hundreds of billions of dollars annually. You can't change the weather, but more accurate predictions will save more lives and property.

I try not to plan my life around the weather, but a few million to possibly offset billions in damage from an incorrect hurricane path prediction is a no-brainer.

Re:Good (0)

Anonymous Coward | about a year ago | (#43772841)

It doesn't promote the "right wing" agenda and therefore only receives funding to lip sync a crisis. The wing nuts have been screaming "big govt - bad" for the past generation resulting in slashing of most budgets outside of defense.

As far as the NWS, I know for a fact that it continues to have "meteorologists" manually check weather devices - even though those manual actions can be replaced by cell communications, people sleep on the job, completely incompetent individuals are simply moved around from station to station - while earning $100K and ultimately receiving life long retirement annuities paid for by you and it.

Citation required. (0)

Anonymous Coward | about a year ago | (#43773073)

Because that's a load of bullshit, except to the extent that private industry is likewise full of the completely incompetent, the irredeemable venal and corrupt, and incompetent executives are simply moved around to a higher pay in a different company and their incompetence covered up (and each one of those earning 150x the average wage of the workers, equivalent to those you claim are on $100K).

Re:Good (1)

kevmatic (1133523) | about a year ago | (#43772933)

Where did you find information on the USA's spending on weather forecasting? Is it really that much lower than that of the European countries?

People seem to see all the embarrassment behind the fact that the European weather forecasting system is so much better, but Europe consists of 50 countries with a total population of 750 million. I don't know how many of those countries put into that weather system funding pot, but I'll betcha its most of them.

The fact that our system, from one country with half the population, is comparable at all seems impressive to me. After all, we're being compared to a continent.

Re:Good (0)

Anonymous Coward | about a year ago | (#43773399)

Yes because to can move buildings off the beach when a hurricane comes, wait er no that's wrong. People when get out of the way when a hurricane arrives instead of having a party, nope thats wrong too.

Sorry this is just a BS give away to IBM and friends. Want to do something about weather related property damage costs? The change building codes, zoning laws and subsidized flood insurance.

Why not collaborate? (2)

applematt84 (1135009) | about a year ago | (#43772689)

I'm a bit confused ... why is so much money being spent if the technology already exists elsewhere? What about remote computing? Why can't we share resources? A 2.6TFlop super computer had better last us a long time. I can't imagine what the "1.21 Gigawatt" power bill will look like.

The fix is easy! (0)

Anonymous Coward | about a year ago | (#43772729)

All you need is a beowulf cluster.

Golden Opportunity! (2)

some old guy (674482) | about a year ago | (#43772747)

Here's a great chance to jump in on another multi-billion dollar government tech boondoggle. Why let SAIC and the other Beltway Bandits scarf up all the big bucks? A bunch of us ought to slap a shell company together and bid like there's no tomorrow. Get on board that gravy train while we can!

If this goes anything like recent FAA, USPS, and VA projects to name but a few, a successful contractor can bill for years while never delivering a finished, operational product.

Surely we can spec a 2.6K TFlop monster, with ancilliary systems, and market it to the GSA purse-holders. Easy math. Calculate the probable actual cost (fair bid price), triple it (IBM, Kray, or SAIC's price), and multiply by .9 = winning bid (never bid too low on a government contact; they automatically chuck out the highest and lowest).

What could possibly go wrong?

Re:Golden Opportunity! (0)

Anonymous Coward | about a year ago | (#43774277)

Your ideas intrigue me and I wish to subscribe to your newsletter.

I guess it's true (2)

WGFCrafty (1062506) | about a year ago | (#43772779)

You don't need a weather man to know which way the wind blows.

Just a 2.6 pflop computer.

Re:I guess it's true (0)

Anonymous Coward | about a year ago | (#43773113)

Installing a window in the forcasters office might be a larger help.

Wow. Moronic comment of the thread right here. (0)

Anonymous Coward | about a year ago | (#43773375)

So lets say that the window points east.

What if the weather is coming from the south east?

Even if the weather is coming from the east, it only gives you a "forecast" a few hours ahead.

Since you want to know in the AM what the weather will be like when you come home, this will be 100% useless to you, even in the unlikely event that the view is appropriate.

So on what way would a window help? AT ALL?

another result of America's priorities. (1)

P-niiice (1703362) | about a year ago | (#43772893)

We need the focus back on public schools, affordable college education, respect for science, and good learning for EVERYONE who wants a good education.

Re:another result of America's priorities. (-1)

Anonymous Coward | about a year ago | (#43772975)

The way to make college affordable is to stop subsidizing it. Artificially increasing demand for a product that is in finite supply can only drive up the cost.

Private schools are universally better than public schools, yet to this day we refuse to learn anything from that observation.

There is no respect for science in government, and never will be.

PROOF AGW IS A SCAM!!! (0)

Anonymous Coward | about a year ago | (#43773025)

Look how this government money is going to computer manufacturers who are in on the scam of AGW so they can get $23mil!

Not only that, but since that computer is being paid out of the weather services' budget, that purchase is going to the researchers too, proving that they're in on the scam because they're getting paid $23mil!

That's $46million dollars for the AGW SCAM RIGHT THERE!!!!

Whereas the Heartland Institute only gets a couple million dollars and Dell only gets a few thousand grand, which is negligible, showing that AGW "Skeptics" are not in it for the money, because that's well over $40million dollars difference right there!!!

(in case this is not obvious, this is sarcasm)

Re:PROOF AGW IS A SCAM!!! (0)

Anonymous Coward | about a year ago | (#43773175)

(in case this is not obvious, this is sarcasm)

It wasn't obvious. It's not that your sarcasm isn't blatant, but that the conspiracy theory denialists are so good at self-parody that it's hard to keep up.

I read something about the forecasting in... (0)

Anonymous Coward | about a year ago | (#43773121)

Nate Silvers book "The signal and the noise". http://www.nytimes.com/2012/09/09/magazine/the-weatherman-is-not-a-moron.html?pagewanted=all

2 cent solution (0)

Anonymous Coward | about a year ago | (#43773131)

maybe it would be cheaper if people would turn off the TV,
go outside and actually look at the sky ... for a weather forecast that is.

How far can you forecast that way? (0)

Anonymous Coward | about a year ago | (#43773823)

An average front will move 5-10knots, you'll be able to see formations of clouds usefully up to maybe 30 miles away.

About a five hour forecast.

But that would only be useful for people within about 10 miles of you.

Therefore you''ll need several thousand people to pop outside and tell everyone else what the weather is likely to be LESS THAN SIX HOURS AHEAD.

It's a 2 cent solution worth every dollar...

Train Wreck Syndrome (1)

jasnw (1913892) | about a year ago | (#43773235)

This is how Government funding works. I was at a workshop on the then-new field of space weather forecasting in the mid 1990s where the keynote address was given by Dr. Joe Friday, at the time the head of the NWS. He pointed out that we would see no serious funding from Congress until there was the space-weather equivalent of a train wreck that kills many voters, or costs the monied interests lots of dinero. (Joe later lost his job when a non-forecastable flood in the mid-west that exceeded the 100-year flood levels wasn't correctly forecast. In this case, the solution was of the Shoot the Messenger variety since the real cause of the bad flooding was lousy planning by the Corps of Engineers.) The local government version of this is not putting a stoplight at a bad intersection until someone, preferably a cute child or pregnant mother, is killed there.

More H1B Visa effects (2, Interesting)

Anonymous Coward | about a year ago | (#43773253)

Posting anon to avoid burning bridges. NCAR has tried to develop better forecast models but they've layed off experienced US staff to hire foreign H1B grad students to write their software. I lost my 18+ yr position as a software engineer at NCAR, while helping to replace the 1980's crap they use to verify the accuracy their models with modern software, using modern techniques . They have great hardware but very amateur software. I got a "we've lost funding for you" while they were hiring H1B's. I was often the only US born software engineer in many of the projects I worked on at NCAR. The US could have much better forecasts, but the public wants everything on the cheap. The Europeans are doing better because they hire professionals to do development and charge for the output. IMO, American weather science is quickly becoming a joke.

Now, can we please upgrade their NEXRAD backhaul? (5, Interesting)

Miamicanes (730264) | about a year ago | (#43773349)

Supercomputing improvements are nice, but I personally want to see them get the cash to profoundly increase their NEXRAD backhaul (the data lines connecting their radar sites to the outside world).

Right now, they're HORRIBLY backhaul-constrained. I believe most/all NEXRAD sites only have 256kbps frame relay to upload raw data to NOAA's datacenter for further processing & distribution to end users. As a result, they're forced to throw away data at the radar site to trim it down to size, and send it via UDP with little/no modern forward error correction. That's a major reason why glitches are common. In theory, the full-resolution data is archived to tape on site and CAN be mailed in if some major weather event happens that might merit future study, but the majority of collected data gets archived to tape, then unceremoniously overwritten a few days later. And most of the tapes that DO get sent in sit in storage for weeks or months before finally getting added to their near-line data archive.

The low backhaul bandwidth is made worse by the fact that the secondary radar products (level 3 radar, plus the derived products like TVS) get derived on site, and wedged into the SAME bandwidth-constrained data stream. That's part of the reason why level 3 data lags by 6-15 minutes... they send the raw level 2 data, and interleave the previous scan's level 3 data into the bandwidth that's left over. I believe the situation with TDWR sites is even worse... I think THEY actually have a single ISDN line, which is why level 2 data from them isn't available to the public at all.

As I understand it, they can't use lossless compression for two reasons -- since they have no error correction for the UDP stream, a glitch would take out a MUCH bigger chunk of data (possibly ruining the remainder of the tilt's data), and the error correction would defeat the size savings from the compression. Apparently, the processors at the site are pretty slow (by modern computer standards), so it would also add significant delay to getting the data out. When you're tracking a tornado running across the countryside at 50-60mph, 30 seconds matters.

If NWS had funding to increase their backhaul to at least T-1 speeds, they could also tweak their scan strategies a bit to make them more useful to others. For example, they could do more frequent tilt-1 scans (the lowest level, which is the one that usually affects people the most directly), and almost immediately upgrade all current NEXRAD sites to have 1-minute updates for tilt 1 (adding about a minute to the time it takes to do a full volume scan, but putting data more immediately useful to end users out much more frequently).

Going a step further, more bandwidth would open the door to a fairly cheap upgrade to the radar arrays themselves... they could mount a second antenna back-to-back with the current one with fixed tilt (ideally at 10cm, like the main one, but possibly 5cm like TWDR if 10cm spectrum isn't available, or a second dish of the proper size for 10cm wouldn't fit), and do some moderate hardware and software tweaks that would effectively increase their tilt-1 scanrate to one every 6-10 seconds (because every full rotation of the main antenna would give them a full tilt-1 rotation off the back). This means they could send out raw tilt-1 data with 6-10 second frequency. It's not quite realtime, but it would be a HUGE improvement over what we have now.

Unfortunately, NWS has lots of bureaucracy, and a slow funding pipeline. I think it's safe to say that the explosion in popularity of personal radar apps, combined with mobile broadband, almost totally caught them by surprise. Ten years ago, very few people outside NWS were calling for large-scale NEXRAD upgrades. Now, with abundant Android and IOS apps & 5mbps+ mobile data the norm, demand is surging.

That said, I hope they DON'T squander a chunk of cash on public datafeed bandwidth instead of upgrading their backhaul. I'd rather see them do the back-end upgrades that only THEY can do, and tell people who want reliable & frequent upgrades to get their data feed through a private mirror service (like allisonhouse or caprockweather) who can upgrade their own backhaul as needed, instead of having to put in funding requests years in advance.

Re:Now, can we please upgrade their NEXRAD backhau (0)

Anonymous Coward | about a year ago | (#43773685)

Sounds like what the Nexrad sites need is a the kind of broadband Internet connection that most households can get. Ok, there is a reliability issue, but using the Internet as a primary and the existing system as a backup might be ok.

Re:Now, can we please upgrade their NEXRAD backhau (2)

JTinMSP (136923) | about a year ago | (#43773873)

One key thing you missed. The NWS 88-D Radar system *can* take a scan every minute at the expense of resolution and distancee. A "full" scan across the commonly used tilts takes *six* minutes. You can have a OC-3 to every radar site, but you're only going to get data ever six minutes most of the time.

Reasons why this is important (3, Informative)

PineHall (206441) | about a year ago | (#43773539)

Cliff Mass, University of Washington Atmospheric Sciences Professor, has been arguing for an upgrade [blogspot.com] for a long time. He sees great potential [blogspot.com] for this new system if used right. The reasons [blogspot.com] for the upgrade boil down to having "huge economic and safety benefits" with better forecasting, and he says these benefits are within our reach.

Wrong answers that much faster (0, Troll)

Anonymous Coward | about a year ago | (#43773813)

Wow...we should be able to get to the wrong conclusion that much faster now. It's not the computer systems that are failing...rather the predicting algorithm the US uses.

OK... (1)

chipperdog (169552) | about a year ago | (#43774241)

Which big government contractor needs work now? That seems to drive these projects more than actual need. I'm guessing the NWS/NOAA has plenty of computing resources, just need to fine tune the models a little bit and collaborate techniques with the Europeans...
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...