Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Calculating a record-breaking 31.4T digits of Pi with GCP

cloud.google.com

Post reply on HN
Print view View original

71–80 of 118 posts

Re: Calculating a record-breaking 31.4T digits of Pi with GCP

#71

Post by rkangel »

megaremote wrote: post #42

Does pi compress? It must if it is only numbers, but only by a half?

If you consider compression to be 'representing information in a way that you can recover with some processing (from less information)', then any programmatic definition of Pi is an enormous compression of the value.

Re: Calculating a record-breaking 31.4T digits of Pi with GCP

#73

Post by rkangel »

s_gourichon wrote: post #22
isostatic wrote: post #20

Earlier quoted context omitted.

The US and Philippines use month-day-year. The rest of the world doesn't. The most popular format is Day-Month-Year, followed by Year-Month-Day.

Yes. ISO chooses year-month-day, which puts largest component first and smallest last. This has the nice benefit that treating it as a string and doing alphanumerical sort matches the actual day sort. Ref. https://en.wikipedia.org/wiki/ISO_8601

The main benefit of Y-M-D, is that no-one uses Y-D-M, and the 'Y' component is easily recognisable. So if you use Y-M-D, then everyone knows it's Y-M-D and there is no ambiguity.

Re: Calculating a record-breaking 31.4T digits of Pi with GCP

#75

Post by Chirono »

ousta wrote: post #70
aaaaaaaaaaab wrote: post #32

Meh, throwing raw power at the problem is not that impressive. Bellard's [1] 2009 record was much more impressive, because he used a clever formula to break the existing record with a mere (albeit beefy) desktop computer: https://bellard.org/pi/pi2700e9/ The record he broke with his desktop PC was made using a supercomputer cluster. [1] If somebody is not familiar with him, he is also the original author of QEMU, ffm…

I struggle to see any complexity into google approach - unlike bellard's one that is an amazing feat. They basically pulled more machine to compute more. nothing really impressive. pretty much any dev with that computing power could have done it

I think that's the point they are trying to make; that doing this with GCP is super boring and easy.

Re: Calculating a record-breaking 31.4T digits of Pi with GCP

#76

Post by aaaaaaaaaaab »

ousta wrote: post #70
aaaaaaaaaaab wrote: post #32

Meh, throwing raw power at the problem is not that impressive. Bellard's [1] 2009 record was much more impressive, because he used a clever formula to break the existing record with a mere (albeit beefy) desktop computer: https://bellard.org/pi/pi2700e9/ The record he broke with his desktop PC was made using a supercomputer cluster. [1] If somebody is not familiar with him, he is also the original author of QEMU, ffm…

I struggle to see any complexity into google approach - unlike bellard's one that is an amazing feat. They basically pulled more machine to compute more. nothing really impressive. pretty much any dev with that computing power could have done it

Exactly. It's a matter of K8s configuration, which a devops intern can do in a few weeks probably.

Compare this with the Chudnovsky brothers, who built their pi calculating "supercomputer" from commodity parts in their NY apartment, back in 1992: https://www.newyorker.com/magazine/1992/03/02/the-mountains-...

(and being mathematicians, they also discovered novel formulas for speeding up their calculation)

Re: Calculating a record-breaking 31.4T digits of Pi with GCP

#78

Post by angrygoat »

Alexander Yee's writeup is interesting – CPU utilisation was only about 12%, they encountered quite nasty I/O bottlenecks (particularly for writes.)

http://www.numberworld.org/blogs/2019_3_14_pi_record/

Contradicts the Google blog a little, especially where he points out that they hit performance issues with live migration (Google said it worked fine without impact on the application.)

Post reply on HN