Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Friday July 31 2015, @10:03PM   Printer-friendly
from the cue-'does-it-run-linux'-jokes dept.

China is planning another petaflop supercomputer, this time to support what will by next year become the world's largest radiotelescope.

The telescope itself, a 500 metre monster that's scooped into a hilltop in Guizhou, has been under construction since 2011.

This week, engineers began installing the 4,450 panels that will make up the FAST (Five hundred metre Aperture Spherical Telescope) facility, which the Middle Kingdom's Academy of Sciences reckons will be able to detect radio signals from more than ten billion light years' distance.

More importantly, its huge size will also mean FAST can pick up even fainter signals than those captured at today's biggest radiotelescope, the [300 metre] Arecibo Radio Observatory in Puerto Rico.

Xinhua reports that the instrument will be supported by Skyeye-1, a petaflop facility that'll connect to FAST with 100 Gbps links.

The Institute of Computing Technology under the Chinese Academy of Sciences(CASICT), Dawning Information Industry Co and China (Guizhou) Skyeye Group will build what's to be called the Qiannan Super Computing Center in Guizhou.

FAST's daily peak demand is predicted to exceed 200 teraflops, with first stage storage of more than 10 petabytes, CASICT researcher Zhang Peiheng told the state-run news agency.

More information at Wikipedia; arXiv.org has the 2011 abstract and full PDF.


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 1) by ese002 on Friday July 31 2015, @11:55PM

    by ese002 (5306) on Friday July 31 2015, @11:55PM (#216577)

    After all these years. Meanwhile, they keep threatening to close the original. It seems the astronomy community believes that money is better spent on other projects. The Chinese dish looks like it will be more aimable, something which is a significant limitation of Arecibo Is this enough to make it worth building? Many seem to think Arecibo is not useful enough to keep *operating*.

  • (Score: 3, Insightful) by bob_super on Saturday August 01 2015, @12:53AM

    by bob_super (1357) on Saturday August 01 2015, @12:53AM (#216595)

    This one has much better tracking capabilities than arecibo. Time on target must be important

  • (Score: 3, Insightful) by Adamsjas on Saturday August 01 2015, @01:01AM

    by Adamsjas (4507) on Saturday August 01 2015, @01:01AM (#216599)

    Arecibo has a lot of problems.

    It is limited in its sky coverage and tracking capabilities due to the ancient technology used to slew the receiver reflector over the canyon. The canyon bowl surface is far from ideal, and not entirely stable. Wind, rain, and low elevation makes it a huge compromise.

    Long baseline arrays are a better, cheaper, more steerable solution.

  • (Score: 2) by wantkitteh on Saturday August 01 2015, @02:00PM

    by wantkitteh (3362) on Saturday August 01 2015, @02:00PM (#216752) Homepage Journal

    Arecibo was built 52 years ago. Nuff said.