Slashdot Mirror
Supercomputers Race to Predict Storms
pillageplunder writes "CNN has an interesting article on how different supercomputers from around the world are working to predict large storms tracks. The 3 days it takes now has been cut in half. Cool read."
Fortan is just as scalable as C. And with modern supercomputing libraries and toolsets, it's probably even better suited to the task. Companies like IBM, SGI, and Intel continue to tune and tweak their Fortran compilers for the latest CPUs (R16K, Pentium 4, Power5, Itanium2, etc).
There are a lot of existing, hightly tuned fortan algorithms out there and plenty of scientists to keep the code running.
And much less free on weather.com. Who decided to do that anyway? Charge for a best-guess on a 5-day forecast? I can see that for free on my rabbit ears on the TV.
Two roads diverged in a wood, and I - I took the one the bus load of girls just went down.
It's easy to think this looking at the paths, but it is not true. Guidance is generally greatly affected by the placement of high-pressure ridges and their future erosions/strengthenings. Frances could have just as easily turned harmlessly to the north had there not been a strong ridge keeping it where it was, and Ivan could have headed east to Mexico had the ridge to the north not eroded. In both cases, they behaved roughly as predicted. The paths of hurricanes are predicted fairly accurately these days, and it is mostly due to these models.
The most difficult part of the job is predicting hurricane intensity, which is not fully understood. That's why Charley caught everyone off-guard when it abruptly strengthend, and similarly in 2002 there was unexpected relief when Lili (who looked a lot like Ivan) weakened overnight just before it hit land.
xkcd.com - a webcomic of mathematics, love, and language.
When your code runs for days, recompile time is pretty trivial. Fortran also has massive speed optimizations over Java. Most code like this has a really small inner loop that runs as fast as it possibly can, and even the smallest of performance hits from things like exceptions, object dereferencing, register loading of potentially aliased variables, and 1001 other minor things that goes on in the background that you don't see can increase run time by hours. Fortran is a least common denominator that lacks the flexibility of programming, but makes up for it by allowing the compiler to do all sorts of tricks like automatic parallelization of these inner loops. Java and C++ don't even come close to this, and hand coding such things is a gamble on efficiency for a good programmer, a sure loss for a mediocre programmer.
One lesson of object oriented is that you should let the language do the work for you. Sometimes this means that you shouldn't use object oriented languages.
Not really. distributing the load like SETI isn't really an option in most of these simulations because of the large amount of comunication that needs to occur between the nodes. Problem is this isn't a bunch of independent processing tasks like in SETI. This is the same reason why fast interconnects like myrinet and infiniband are often used to connect the nodes in the clusters.
- dharhas
Fortran compilers are guaranteed that the programs do not try to do strange things behind their backs (such as pointer aliasing). Therefore they can make optimizations that would be almost impossible to prove valid in, say, C. Also, Fortran numerical libraries are of very high quality.
Finally! A year of moderation! Ready for 2019?
Im guessing the choke point would be the drives read speed. I wonder if they run seti@home on these things. :-)
Support more choices in goverment-Vote 3rd party.
Oh, that's good. Screw around with mother nature some more.
Did you ever consider that things happen for a reason? Like balancing global heat loads and adjusting the water vapor cycle?
So instead of having a bunch of light to heavy storms, we'll end up with having ONE BIG MONSTER that we *can't* stop.
Big difference between casual observer and what the people doing the modelling are doing ;) To predict the EXACT track/intesnsity/storm surge/rainfail is going to be impossible...there are just way too many factors that determine what any particular storm will do at any given time. Same thing with earthquakes, flashfloods, tornadoes, etc. Have you read any of the Discussions on the National Hurrican Center's website for storms? They have many different models and then try to figure out which one is more accurate...and it's not where the storm was or is that's important, but what's going on in the atmosphere all around it... it's not something simple. And it's hard to make accurate predictions with limited data. And the predictions are getting better... I'd imagine the area of coastline that got smaked by Charley was under a hurricane warning at the time of landfall, right?
Everyone that lives in hurricane prone areas SHOULD be prepared for this type of thing. I'd rather be overly cautious with a major hurricane barrelling down on me... If you don't like dealing with the possibility of tropical cyclones, move inland :)
Charley swerved just before landfall, Frances stopped dead in the water 60 miles off Florida, and Ivan "bounced" off Jamaica, shifting its path by 500 miles, none of which were predicted. Possibly, none of which are predictable. If you can't warn people where landfall will occur when it takes some non-obvious path, then what's really the point?
Weather simulation is not a tast you can cut up into a bunch of smaller tasks and farm out. If you cut the atmosphere up into lots of little chunks to model, after each step every chunk needs to know what the results from all the chunks around it where.
If you're waiting for those results at Internet (Latency: 100ms) speeds instead of intra-system speeds (latencyL 1 us) it takes you 100,000 times as long to get your data.
With SETI, all you do is get the data once, compute, and send back the answer.
paintball
This sort of code is probably heavily matrix oriented. Fortran handles matrix operations, making matrix operations as easy to program as simple arithmetic.
As others have already noted, this isn't a SETI-type problem. Whereas SETI has a large amount of data that must be parsed, analyized, etc. individually, a CFD simulation requires inter-node communication, meaning each portion of the solution that is running in parallel must be able to communicate with other portions. For a distributed environment, sometimes things like gigabit ethernet just aren't fast enough--hence the market for myrinet and infininet. Many times for a complex problem bandwidth is more of an issue than processor muscle.
bueller...bueller...bueller
If you think about it, 3 days notice is not enough to have every person in a metropolis patch up their houses and move to higher ground. Some might say that everyone with the possibilty of getting hit by the storm should prepare, but imagine having to board your windows every 3 weeks or so only to be missed by the storm.
The boarding-up problem can probably be simplified with rigid mount points and locks on pre-fitted panels. I'm sure a solution can be designed for second story windows where it can be installed and locked in from the inside of the house and still be on the outside to protect the glass. I imagine it could be done such that it only takes a couple minutes per window, five for the largest ones and you can be packing within an hour or two. That might not help for situations where there are just too many people on the road to evacuate in time.
I really can't speak to the other issues raised.
Mother nature is going to do whatever the @#$%@#$ she wants to do, no matter how hard we try to say what we think she is going to do. All credit is due, however, to the people who by making their best guesses with the equipment and data they have ARE SAVING LIVES. Good on them, I say!
"...Two roads diverged in a wood, and I-- I took the one less traveled by, And that has made all the difference."