Slashdot Mirror
Sun Puts Data Center Through 6.7 Earthquake
An anonymous reader sent in a video clip showing Sun experimenting with shoving a data center through a simulated 6.7 Earthquake.
Everything stays running, but some power cords came out and some screws worked loose. It's still kind of neat to see a bunch of racks shake like a polaroid.
thats what sun is spending money on before its taken over?
What hard drive survived that, that's what I'd like to know.
FUCKING HELL is that an embedded video I see in the story!? Holy shit, the geek website is ... in step with the times?
The real link to the video is in the firehose related article
Perception is the thin dividing line between reality and fiction.
Eh?
Sweet, a link in a summary to the summary itself. Just what I've always wanted!
Their may be a grammatical error, misspeling, or evn a typo in this post.
Only on slashdot does this refer to server hardware.
(At Hooters, it refers to server software).
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
(At Hooters, it refers to server software).
Server firmware, please. Typically, embodied in silicon(e).
Welcome to the Panopticon. Used to be a prison, now it's your home.
after the quake, it's still running Solaris.
of outkast, many many many times better than the black eyed peas.
He tried to kill me with a forklift!
...it's so simple, beautiful, elegant, yet sometimes I wonder wtf its doing.
Solid State I'm assuming????
The above comments are not guaranteed to make sense to anyone other than the author...
I knew there was something familiar about this. I stumbled upon it on a slow day at work a couple of years ago. The video is dated 2007 at the end.
When they can have this type of earthquake and not have any IO errors from the disks nor do any tapes fall off the walls of the inside of the tape library, then I'll call this a success. As someone that has had to retrieve a tape that was dropped by the robot of an old STK "Powderhorn", this would be a pain.
Everyone has pictures of racks sliding across the room and CRT terminals dangling from desktops. The surprising thing was how much rebooted immediately after the power returned. And even in that year the pre-web internet was more reliable than the phone company. Email worked better than many phones.
Whatever happened to giving units along with numbers, so the numbers will actually mean somthing? I'm assuming Richter here, but as we all know assumption is the mother of all screw ups.
Pretty good is actually pretty bad.
I think the link in the article has been slashdotted.
Everything stays running, but some power cords came out, some screws worked loose
How did the servers keep running without the power cords? Or do they refer to power cords on the UPS's?
For that matter, since when does /. have video in TFS?!? Try the link... it points to itself!
It is dangerous to be right when the government is wrong.
Everything stayed running... the failures consisted or power cords coming out
So by "running" I think they mean "didn't break"
Is it just me, or does the video indicate this was recorded January 17th, 1994?
Slashdot: Playing Favorites Since 1997
Energy release by Mw6.7 Earthquake is as following.
6.7 16.2 megatons (TNT) 67.9 PJ (Joule equivalent)
See more here.
http://en.wikipedia.org/wiki/Richter_magnitude_scale
Most things can survive Mw6.7 earthquake just fine. The question is, can it survive Mw7.0 earthquake or bigger. A earthquake that is Mw6.7 is just strong.
my manager sees this shit.
now whenever i mention colocation and its impending budget, ill have this godforsaken thing thrown in my face. important facts like "way outside its normal envelope" will fall to the wayside as superbox 9000 will solve all the companies woes, cause our shareholders to sing, and increase productivity by big number!
then, when i integrate it with both the cloud and the grid infrastructure, ill see a completely service oriented architecture designed to leverage our aging, proprietary, uncompetitive, lazy, and barely piece of suck ass assets to rocket us in direct competition with google, before we overtake microsoft!
glad they tested it, but sad it was never really emphasized the box shouldn't be guaranteed to specifications that resemble porn.
Good people go to bed earlier.
How did the servers keep running without the power cords?
I got 3 words for you - redundant power supplies.
Although I don't think this is the case here. If you RTFA (or is it WTFV?) they do say that "the only failures were associated with power cords coming loose" - so they do consider this aspect to be a failure - but its a very easily fixable one - many servers have ways to fix power cords to the chassis of the server in a way that you cannot yank the power cord out easily.
-Em
RelevantElephants: A Somatic WebComic...
I dunno, maybe only SOME of the power cords came out? But, I think the summary was supposed to illustrate that the servers didn't fail because they broke, even though they might have 'failed' because the power cords came out.
They need to test if Data Centers can survive a Myth Busters taping. Thats a REAL test.
http://www.kcra.com/cnn-news/19016582/detail.html
"Capital punishment makes the state into a murderer. Imprisonment makes the state into a gay dungeon-master"
I designed a navigation display product some years ago for shipborne application (think bridge of supertankers) and we put it through a standard shake & vibe test. Everything came through fine except the video was scruzled. At first we assumed the CRTs died but upon investigation we found that the connectors on the MB ate their way through the gold fingers on the PCI video card. As electrical engineers we learned a lot of hard lessons. Shake and vibe are tough and since every system is going to have different harmonics it is hard to generalize. Simple rules of thumb and intuition may serve you poorly. In some cases shock mounts made things worse.
Anonymous just because I'm to lazy to login...
The way I read it, is that the data center as a whole stayed up and functional. I'm sure it's built with enough redundancy to maintain service through a failure of a few machines/drives/switches/etc...
Not every power cord came loose, the "system" compensated, and the box kept on serving.
Now they need to test what happens when the field tech is replacing a drive right when the earthquake hits. That should be some fun watching! Does he still get the drive replaced?
Let's find out!
Every once in a while, editors get to approve the best things...
http://www.collude.biz - Ignore this, it's for Project Honey Pot.
Real computers have more than one power cord.
... will it blend?
Everything stayed running... the failures consisted or power cords coming out
So by "running" I think they mean "didn't break"
Redundant power cables? Although to be fair, in a real data center, KVM pushcarts and jewel cases left in partially filled racks would be a big factor in causing wire damage. Not to mention server mounting arms extending.
--- Justin Dearing http://www.justaprogrammer.net/ We're just programmers.
and you get rickrolled. Is that better?
I noticed the racks swaying quite a lot during the test probably because they were bolted down to the floor as a standard procedure with data centers. I think that for the next design of their "black box" system they should also bolt them to the ceiling to prevent the swaying of the racks and this would probably also solve the grill damage problem seen on the top of the rack along with structural damage like buckling, bending, twisting in the body of the server racks.
The hard piping issue of using copper tube pipes for heat transfer with glycol or water might pose another problem, luckily copper is more malleable than other piping alternatives so the pipes would likely bend and deform easier before breaking or shearing due to the random motions.
This little problem of cables falling out happens in regular data centers all the time due to heat expansion/contraction creep, loose connectors, or accidental unplugging is an old problem but an easy one to fix. A little loop connector or cable-tie for the power cable mounted to the back of the case as seen on Compaq and HP server systems would take care of any cables falling out of the receptacles.
Wow, what a flash back in time...I watched this video at least a year or two ago....
Indeedy--the © notice at the end of the video is 2007! (Yeah, I really did WTFV.)
I'd put money on them using redundant power supplies for starters, and the way they were speaking it sounds like they took the survival of some/most of the machines as 'working'. Which to be fair is more than you'd expect if your data center got that kind of shaking.
...who has no idea what "shake like a polaroid" means? Last I knew, a polaroid was a instant camera using chemical-based films, and was not intimately connected with geological stressing.
Can someone please demonstrate what a shaking polaroid looks like so I don't feel like I'm missing out on hacker lingo.
Moderation Total: -1 Troll, +3 Goat
Lol what about the poor SOB who's in that tin can doing maintenance on the server when the earthquake hits.
..just because you can, doens't mean you should...
as I was trying to figure out how to actually see the video and clicking on the link like an idiot just to get back to the same place, I figured out that a large blank space below the text is where the video is supposed to be. Since it seems it is slashdotted I cannot see it. It would be better if you could just give the link in the text so I can save it and look later and/or at least say that the video is below.
No... in a real datacenter, there's no shit cluttering up the floor or the racks, and KVMs are a permanent solution -- either with a console in every rack, or a centralized solution. Of course, if you shop with Sun, there's really no need for a KVM because their ILOM rocks so hard -- run a little java applet, and you've instantly got keyboard, mouse, and you can mount local disks / images onto the machine. And, if you can VPN / tunnel, you can do it securely from anywhere in the world. In my experience, Sun salespeople are bastards, but their hardware is incredible.
There's no room for a KVM pushcart in the Sun Modular Datacenter.
All sun servers, new ones anyway, have an integrated lights out management card (ILOM) that you would use instead of a KVM switch. It allows you to connect to the server, even if it's powered off.
If you were putting this in a seismic zone I would assume you would install some rack drawers if you would have small objects such as jewel cases so you wouldn't have them just laying around. The design of the unit doesn'nt seem to have any shelves, or things you could use as a shelf to put these items on anyway.
The racks are put in the container sideways and there is a side panel. The rack slides out into the aisle if you need to service anything in the rack.
I posted a link to this video on the previous thread on the new Internet Archive Data Center that used one of these modular data centers. I guess someone found it interesting and didn't notice how old it was.
Dual Opteron < $600
Sure, the servers survived the quake, but what of the datancenter itself ? I would not be surprised if THEIR power mains or network uplinks went to shit after such a rumble.
-Billco, Fnarg.com
I saw this at least 3-4 years ago.
According to this article on the modular datacenter it seems like they started doing better after they changed their name. Guess even a big company needs to worry about SEO? :)
The coolest (literally I guess) was where the put 50 of these in an abandoned mine in Japan. The mine had a constant, low temperature which would reduce cooling costs.
They mention 4 other customers in that article which is over a year old. There are probably more.
The nice thing is that you can also get Sun remote monitoring of the boxes. So if they detect a fault, they can alert a local non claustrophobic service provider to swap out a drive or whatever. Though your IT staff might miss having a NASA like command and control center.
Dual Opteron < $600
How a 7, 8, or even 9? :D
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
I can't believe it. So a bunch of corporate suckers emulate an earthquake to advance their technology? In the core of Silicon Valley? Where is Spiderman?
"Hey Ya" by OutKast. Best pop song of the new millennium, haven't found anyone yet who dislikes it.
(Apart from the freaks on here who I'm sure would love the chance to dislike)
If this were really happening, what would you think?
I just ran across this link that shows where sun modular datacenters have been deployed. Looks like a couple dozen. Dosn't give much information on them thouhg.
Dual Opteron < $600
There's no room for a KVM pushcart in the Sun Modular Datacenter.
Can you tell me what's a KVM pushcart, please?
Searching in google doesn't seem to turn up anything.
Thank you.
A KVM is a device that allows you to use one one keyboard, video display and mouse and switch across multiple computers.
A pushcart is a frame on wheels that you can put stuff on and push around.
A KVM pushcart is a KVM switch, monitor, keyboard and mouse on a cart that you can push around to bring to different racks to use the KVM without having to have a KVM setup in each rack with dedicated connections to each server.
Dual Opteron < $600
Sure, the servers survived the quake, but what of the datancenter itself ? I would not be surprised if THEIR power mains or network uplinks went to shit after such a rumble.
The box is the center. The only hookups are hot and water lines, ethernet and power.
Dual Opteron < $600
...what they mean with Solaris on Sun being super-stable? You can shake it and it'll still run? On a more serious note... I heard about the blackbox before; I also recall reading something about other vendors doing a similar thing. I thought the almighty Google used containers in some places as well. Is there any data about how the competition is handling stuff like this? Heck, it might even be the start of a my-dick-is-bigger-than-your-dick list similar to the top 500 supercomputers list: which machine(s) can handle the most abuse?
/var/run/twitter.sock is a twitter socket puppet.
When will Outkast learn:
"Do not shake or bend developing pictures."
http://www.polaroid.com/service/userguides/photographic/one600classic_ug.pdf
or
http://74.125.93.104/search?q=cache:TMzxjpjAo28J:www.polaroid.com/service/userguides/photographic/one600classic_ug.pdf+do+not+shake+site:polaroid.com&cd=1&hl=en&ct=clnk&gl=us&client=firefox-a
Go back to MySpace and stay there until you learn how to speak better than a awful pop video.
MORE THAN ONE powercord? Lordie lordie. How will Rand ever stop the Wheel of Time now?
I didn't know there's such a thing. Thank you for your explanation!
Yes, Tri-X lives on, available for 55 yrs+ in 135 format. and for color reversal prints (direct positive process), Ilford CibaChrome continues under the name Ilfochrome Classic. You can still make ArT and it is still pricey.
Please Earth Quake !! What will happen to the Junks surrounding this black Box.... Its seems look a like 10ft container .... What happen if another black box by side of this black box hit one to one...... The impact logically will be more higher than 6.7 magnitude earth quake impact.