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

My £60 ARM server

ewanleith.com

Post reply on HN
Print view View original

61–70 of 72 posts

Re: My £60 ARM server

#61

Post by sciurus »

runeks wrote: post #24

"[...] ARM servers are a big part of the future [...]" Has anyone actually compared x86 and ARM doing some sort of comparable task, and measured how much power they both consume performing the same task? I know ARM uses a lot less power than x86, but if it uses 10 times as little power but takes 20 times as long to do the same task as an x86 server chip, it still uses more power. I'm not saying it isn't more power ef…

See Google's paper 'Brawny cores still beat wimpy cores, most of the time'. It concludes that 'Slower but energy efficient “wimpy” cores only win for general workloads if their single-core speed is reasonably close to that of mid-range “brawny” cores.'

http://research.google.com/pubs/archive/36448.pdf

HTML version at http://bit.ly/KOhENi

Re: My £60 ARM server

#62

Post by ajross »

runeks wrote: post #24

"[...] ARM servers are a big part of the future [...]" Has anyone actually compared x86 and ARM doing some sort of comparable task, and measured how much power they both consume performing the same task? I know ARM uses a lot less power than x86, but if it uses 10 times as little power but takes 20 times as long to do the same task as an x86 server chip, it still uses more power. I'm not saying it isn't more power ef…

ARM boards are generally more power efficient for any given compute unit you want to benchmark, but it's not a huge advantage

And note that there are Intel solutions in the "cheap, low power" market too. There are ~$80 Cedar Trail boards (I forget the part numbers off-hand) which take 2GB of DDR3 memory and run on 15W or so at full CPU utilization.

Re: My £60 ARM server

#63

Post by xxiao »

use Marvell's sheevaplug is better off for this. beaglebone is really for industrial control, beagleboard are for generic purpose while pandaboard is for mobile computing. SD cards are made for consumer class cameras, but you can buy industrial level SLC SD, for $50 per 4GB, in that case I will just go with SSD.

Re: My £60 ARM server

#65

Post by mbell »

rwmj wrote: post #45
rythie wrote: post #27

If it had a lot more memory that would be interesting. Though really you can get a 256Mb VPS for £5/month http://prgmr.com/xen/ and you don't have a setup charge or unreliable hardware.

ARM has strange limits on the amount of memory. The Versatile architecture supports up to 256 MB max. The Versatile Express architecture has a 2GB limit (and no PCI!) Then you have to remember that ARM is presently 32 bit, so you're going to have trouble going over 4 GB at all, which is currently a very small amount for a serious server. 64 bit ARM isn't going to be widely available for at least another 2 years, and…

The Cortex-A15 can address 1TB of ram, its still 32bit so there is the requisite per process limit. Last I heard Cortex-A15 SoC's will be available in volume this year.

Re: My £60 ARM server

#66

Post by videmsky »

rmoriz wrote: post #64

I build a couple of ARM "servers" out of a PandaBoard (dual core + 1 GB RAM) and a self-made case (lasercutter) http://www.flickr.com/photos/rmoriz/sets/72157629535790241 http://www.thingiverse.com/thing:18645 They're running Linaro, the Ubuntu-ARM-development-fork by ARM/Linaro.org

Whoa! It's like a micro-server and a makerbot mated. Nice!

Re: My £60 ARM server

#68

Post by rmoriz »

videmsky wrote: post #66
rmoriz wrote: post #64

I build a couple of ARM "servers" out of a PandaBoard (dual core + 1 GB RAM) and a self-made case (lasercutter) http://www.flickr.com/photos/rmoriz/sets/72157629535790241 http://www.thingiverse.com/thing:18645 They're running Linaro, the Ubuntu-ARM-development-fork by ARM/Linaro.org

Whoa! It's like a micro-server and a makerbot mated. Nice!

I even made a cheesy video ;-) http://www.youtube.com/watch?v=vX61U_q6ksg

Board: http://pandaboard.org/

OS: http://www.linaro.org/

Re: My £60 ARM server

#69

Post by throwaway64 »

m_eiman wrote: post #55
rwmj wrote: post #12

Personally I wouldn't trust a microSD card for any production role beyond being a bootloader. You'll be paged one day only to find you had to make a time-consuming visit to your co-lo, and that all your data was irretrievably gone. The solution to this is to have one "real" PC acting as a NAS and serving iSCSI. You would have to divide the extra cost of this PC across all the ARM servers you have.

You could also get an "industrial grade" SD card; they claim a lifetime of years instead of months. They're usually using SLC Flash and (possibly) more advanced wear leveling compared to consumer cards.

i wouldn't count on that, google's study of hard drive lifetime/durability indicated that "enterprise" hard drives had almost identical lifetimes and performance characteristics to consumer gear (baring of course stuff like 15k RPM drives).

In other words, "enterprise" is more or less exactly the same stuff, but at a premium. Price discrimination strikes again.

Re: My £60 ARM server

#70

Post by itsnotlupus »

nico_h wrote: post #31
jgrahamc wrote: post #14

Earlier quoted context omitted.

Huh? I browse constantly with third party cookies disabled in Safari and Chrome and never have a problem. I can see that site perfectly. And that's certainly not how CloudFlare decides that your machine is likely to be infected. What are you seeing when you visit the site?

I was browsing with firefox with 3rd party cookies disabled, and I saw this message a few times. What irks me and triggered my kvetching is that I "played the game" and did submitted the captcha 3 times and it still didn't work. Twice in Firefox and once in Safari in private browsing mode. Here is a screenshot, with my I.P. address at the bottom: http://imgur.com/n9xFG It looks scam-y too with the "google pays me $x…

You are not seeing that page because your cookies are blocked.

You are seeing that page because CloudFare believes your IP address is behaving badly, for whatever reasons. ( see https://www.google.com/webhp#q=202.72.107.83 )

Instead of outright blocking all traffic from known bad IP addresses, they have a mechanism to let actual users go in. That mechanism relies on a captcha flow, and on setting a cookie in the user's browser to bypass the IP block.

Disclaimer: I am inferring all of this from your screenshot. CloudFare's actual process and intent may vary.

Post reply on HN