Slashdot Mirror
US Reveals Details of $500 Million Supercomputer (nytimes.com)
An anonymous reader quotes a report from The New York Times: The Department of Energy disclosed details on Monday of one of the most expensive computers being built: a $500 million machine based on Intel and Cray technology that may become crucial in a high-stakes technology race between the United States and China (Warning: source may be paywalled; alternative source). The supercomputer, called Aurora, is a retooling of a development effort first announced in 2015 and is scheduled to be delivered to the Argonne National Laboratory near Chicago in 2021. Lab officials predict it will be the first American machine to reach a milestone called "exascale" performance, surpassing a quintillion calculations per second. That's roughly seven times the speed rating of the most powerful system built to date, or 1,000 times faster than the first "petascale" systems that began arriving in 2008. Backers hope the new machines will let researchers create significantly more accurate simulations of phenomena such as drug responses, climate changes, the inner workings of combustion engines and solar panels.
Aurora, which far exceeds the $200 million price for Summit, represents a record government contract for Intel and a test of its continued leadership in supercomputers. The Silicon Valley giant's popular processors -- the calculating engine for nearly all personal computers and server systems -- power most such machines. But additional accelerator chips are considered essential to reach the very highest speeds, and its rival Nvidia has built a sizable business adapting chips first used with video games for use in supercomputers. The version of Aurora announced in 2015 was based on an Intel accelerator chip that the company later discontinued. A revised plan to seek more ambitious performance targets was announced two years later. Features discussed on Monday include unreleased Intel accelerator chips, a version of its standard Xeon processor, new memory and communications technology and a design that packages chips on top of each other to save space and power.
Aurora, which far exceeds the $200 million price for Summit, represents a record government contract for Intel and a test of its continued leadership in supercomputers. The Silicon Valley giant's popular processors -- the calculating engine for nearly all personal computers and server systems -- power most such machines. But additional accelerator chips are considered essential to reach the very highest speeds, and its rival Nvidia has built a sizable business adapting chips first used with video games for use in supercomputers. The version of Aurora announced in 2015 was based on an Intel accelerator chip that the company later discontinued. A revised plan to seek more ambitious performance targets was announced two years later. Features discussed on Monday include unreleased Intel accelerator chips, a version of its standard Xeon processor, new memory and communications technology and a design that packages chips on top of each other to save space and power.
The new fad is a super computing race! Who needs to go to space when you can simulate it in your back yard.
But not before it reports the answer:
42
When Fascism comes to America, it will call itself Anti-Fascism, and tell you to give up your guns.
The ability to work with a larger dataset in "realtime" means better data trend resolution when you're simulating something and more confidence in the result. Scaling-down / iterating is possible but can introduce artifacts.
So you're 1/2 right. Data superset > data subset.
Right before it was going to report the answer, Windows 10 decided it was time for an update. Once installed, the update broke all audio output. Now we'll never know!
Climate change? Solar panels? This project is as good as cancelled.
I don't know.. Super Computers are key in current state of the art nuclear testing, when you cannot just dig a hole and set your latest weapons off underground due to the nuclear test ban treaty. I got a feeling Trump will let this climate change study slide to get better weapons. I call it a win win myself.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
Larger computers allow models to be run faster so you can repeat more runs to see what the probable impacts the model predicts. They also allow finer scale detail in the model and the modellers to add new variables to the model. The amount of exponentially goes up massively if you are mapping the climate and you decrease the size of your grid element (a block where weather conditions are the same). New data is coming in from the field all of the time and our understanding how systems work is always improving. With better computers more variables may be added to a model to make it more accurate. Those variables may have used up too much computer time on the older systems.
I'm not ruling out that there's some national pride at work and the awarding of the contract to a specific company might not have helped out someone.
It runs at an expected 1 exaflops but with the Spectre prevention it'll be running at just under 900 petaflops.