Slashdot Mirror
Interchangeable Data Storage Bricks?
shokk writes "EWeek is reporting that IBM is working on a concept called Ice Cube Storage Bricks that uses a conductive ceramic or mylar plate to transmit data between bricks across an air gap. Research center staff member Robert Gardner says that the idea is 'to walk up to the system, attach the storage and then walk away.' No mention is made of what happens when a brick in the middle of the cube needs to be replaced and the whole thing needs to be disassembled. To be really effective, this would need to be teamed up with some sort of a backplane, but the tech is new and neat."
Well, except for where it was specifically mentioned in the article.
It's getting bad when the person submitting the story doesn't even RTFA."Da ist ein Technölüst in mein Unterpanten!"
More energon!
If you have to replace a block in the middle and the pile collapses, does the server crash and you lose?
Reviews with a twist! http://www.sardonicbastard.com
Now that is some gangsta storage up in here! Props to big blue, fo shizzle. Peace. Out.
Jeysos, u ver en sucj e hury, u gott et al wronng!
"EWeek is reporting that IBM is working on a concept called Ice Cube Storage Bricks that uses a conductive ceramic or mylar plate to transmit data between bricks across an air gap. "
Kind of like a neuron.
I knew that playing with legos would come in handy sooner or later.
I Am My Own Worst Enemy
How long before we see sites dedicated to storage array building contests?
No mention is made of what happens when a brick in the middle of the cube needs to be replaced
Or, when the ice melts?
I don't need a signature.
Did anyone else immediately think of the Superman's Fortress of Solitude? Guys?...Guys?..No?
Graffiti Array Contests
storage tank.. /. article about a storage brick as well, but I can't find it.
I recall another
That name for the individual bricks, coupled with the fact the picture they have on the website of the partially constructed collection looks kinda like this is rather disturbing.
I Am My Own Worst Enemy
the term "air borne viruses".
An Indian-American Hindu committed to non-violent thought/speech/action alarmed by the global explosion of radical Islam
My mind is going.... dum dee dum. B.
Every experiment which ends in a big bang is a good experiment.
Call me a luddite, but I find the hype around "new and neat" technologies a bit worrying. To me, the obvious problem of having to disassemble an entire block of these things just to get to one failed device is an indication of flawed design.
It's a neat idea just like hotswapping was, but it's going to be a while before it's affordable and reliable. I'll wait for that, I think. Until then, I'll just try to imagine a beowulf cluster of these things and stick to my tried and true server setup, sans bricks.
A Proud Member of the Reality Oriented Community.
Imagine a beowulf cluster of these things!
It's Friday. You ain't got no job. You ain't got shit to do.
Caveat: I'm an employee of IBM's Storage Systems/Technology group, but I'm not working on that particular project. I am only discussing things that were in the previous press releases about this product so you won't get anything confidential out of this post.
i c_storage/CIB_Hardware/
The original intent, when this was previewed a year or so ago, was that dead bricks would just stay in there and not require disassembly. See http://www.almaden.ibm.com/StorageSystems/autonom
for some more discussion.
The concern I have (my role in storage systems is error isolation and recovery) is that when you are running all these individual cubes, each one is trying to isolate what might have happened to its peers (or to itself) and when an error starts to propagate from one cube to the next, which it will invariably do sometime, you could end up with multiple cubes saying "IT'S THAT GUY!" and shooting him (ie, cutting him off) when in fact it was yet ANOTHER cube that started the whole thing by corrupting a message and is innocently sitting there not showing any failures.
So assuming that situation occurs, you have 1 failed and 1 not-failed cube which need to be fixed, and shutting off the failed one requires removal, which isn't part of the service model for the product. Needless to say, I'm going to be REALLY impressed when they get this working. My peers at IBM are awesome when it comes to storage, so I'm actually not being sarcastic when I say that.
I remember reading about this a year or so ago.... ON SLASHDOT....
"The two most abundant elements in the universe are hydrogen and stupidity." -Harlan Ellison
I like to call them disks.
A simple RAID 5 type system would be able to mitigate the potential loss of a tray of bricks while the dead brick was swapped.
A more intelligent system may be able to actively monitor the health of each brick, detect a failure, shuffle data around, and plot a path on a screen to the dead brick.
ie. "To replace brick 1234, please remove brick 2345 first. Then remove 4532. Then remove 9786. Then remove 4575. Now remove 1234, replace. Now reinsert previously removed bricks"
Then the system would shuffle the data back to the bricks that went offline.
This idea is really only feasable for "small cubes", less than 5x5 or so... huge grids would need something along the lines of a tray system instead.
So it's just like a fancy RAID setup. Very Borg-like ;)
:(
What the advantage storage-wise? If it spreads the data across all the bricks wouldn't you lose a lot of storage space? I guess the point is ease of use.
The water cooling would be interesting to see, especially for the center Brick in a pile. That also defeats the purpose of ease of use if the center brick fails.
I don't have any experience with RAID. I'm too poor
My interchangable data storage bricks are my computers.
No mention is made of what happens when a brick in the middle of the cube needs to be replaced and the whole thing needs to be disassembled
:)
IIRC, their idea is that when a brick dies, you just leave it there. Imagine a big room as a circle. You build bricks around the circle starting in one corner of the room. As you "upgrade", you stick new bricks on to one end. If a brick dies, screw it, just stick another one on at the end if you need it. When you run out of space at the end of the circle, you start dismantling the old obsolete (and possibly dead) bricks from the starting end and add the new snazzy bricks onto the end.
It's pretty sweet, and that ignores the watercooling channels that require special plumbing
--------
It's OK to be social, just don't tell anyone about it.
I for one welcome our cube storage overlords.
interesting...
"No mention is made of what happens when a brick in the middle of the cube needs to be replaced and the whole thing needs to be disassembled." - http://www.hasbro.com/jenga/
This is stupid.
This comes from a research lab where people have too much money and too many signs everywhere that say, "Think of a new idea." Current storage solutions aren't so broken that they need this kind of fix. If you want to be so frickin clever why don't you work on some high density, cost effective way to eliminate moving parts from storage.
...nother brick in the mass storage unit.
Darn. Doesn't scan.
IBM Makes Progress with Storage 'Bricks'
By John Pallatto
December 16, 2004
IBM has made progress over the past year in developing a new water-cooled, modular mass storage system designed to be highly fault-tolerant and make more efficient use of electric power and cooling capacity.
Called CIB (Collective Intelligent Bricks), the storage system is under development at IBM's Almaden Research Center in San Jose, Calif. IBM officials discussed its work on the prototype intelligent brick storage system with gave reporters
and editors Wednesday as part of a general briefing on its storage research efforts.
CIB is an effort to make highly reliable storage systems from less-reliable standard components, said Robert Gardner, a research center staff member and co-leader of the development project at IBM. The storage units are literally designed as square bricks that can be assembled into large, Rubik's Cube-like blocks.
Each brick has its own CPU, memory, cache and networking connections. This makes the brick "appliance-like and easy to add by end-users," Gardner said.
Individual bricks can have varying amounts of storage capacity of up to 80 GB. The bricks can be assembled into systems containing terabytes or even petabytes of storage capacity.
PointerClick here to read about IBM's recent update of its TotalStorage SAN software suite.
Rather than using typical wire prongs or plugs, the bricks are connected with a
novel technology called "capacitive coupling," in which one block is mated to the next through a conductive plate. Gardner displayed two different prototype couplers, one made of Mylar and the other of thin ceramic. The couplers are actually able to transmit data through the extremely thin layer of air between one brick and the next., Gardner said.
One of the key goals is to make storage systems that are easier to build and maintain by customers, Gardner said. It should be easy enough, he said, to enable a data center technician "to walk up to the system, attach the storage and then walk away."
IBM also believes the bricks will allow it to design storage systems that are as much as three orders of magnitude more scalable than existing storage arrays while reducing complexity and simplifying maintenance, he said.
The bricks can use cheaper, less reliable components because the failure of a single brick or even several bricks will not shut down or corrupt data in the other bricks, because the data is mirrored in other sections of the array or in backup systems, he said. As a result, defective bricks can stay in place until they
are replaced as part of scheduled maintenance, Gardner said.
PointerTo read about how the latest storage systems are gaining built-in grid computing options, click here.
The bricks can be assembled "in a big pile of bricks or it could be a one-dimensional wall of bricks," which could make maintenance even easier. IBM is studying which configurations would be most effective for maintenance, Gardner said.
IBM is experimenting with water cooling with these systems because it is becoming increasingly difficult to provide efficient air cooling for the huge volume of storage devices that customers are cramming into their data centers, Gardner said. He and other IBM officials declined to say when the experimental IceCube technology might be released as a product.
Gardner indicated that his task was to demonstrate the effectiveness and practicality of the system before the company decided whether to release it as a product.
The system offers significant potential benefits to customers, he said, because
liquid cooling can save as much as seven times the amount of floor space required for effective air cooling and reduce the amount of power used for cooling by 20 percent to 50 percent, he said.
The laboratory model displayed by IBM used aluminum cooling jackets to circulate water through stacks of individual bricks. The water would pass through an external heat exchanger on a building roof or in an outdoor tank, much like an air-conditioning system, he said.
You can already fit about 2TB is a large desktop case. These cubes only store 60GB/cube.
I would rather use loads of desktops, each with a local RAID array. Depending on bandwidth needs, I would either connect them to a common gigabit ethernet router (not so scalable) or set up dedicated routers in a tree heirarchy with larger and larger pipes as you get near the root.
Scalability should not be too much of an issue, and with 10 or so HDDs in a single case, you don't waste too much electricity.
Naturally, they would be running Linux.
Both sound like lego bricks shoved into slots in a backplane (;-))
--dave
davecb@spamcop.net
+2 Informative
If I had patented this idea when I had it ten years ago I would have been rich! (Let's for the moment gloss over the fact that I'm too dumb to design a working prototype.)
I think they should get Frank Gehry working on some new sculptural/architectural designs with this technology, so companies can go back to the good old days of showing off their data center to the world with a huge wall of glass. Aww, screw it, just cover the whole thing with titanium plates. Same effect.
Ask yourself how biological systems handle error propogation?
If we are lucky it'll look like the Gibson in Hackers complete with arcing.
In addition to storing corporate data, they play rap music and scratch records.
When are they coming out with Snoop Dogg Storage Bricks?
I think that the real invention here is not the drive array itself, but the connector that is used. This would be a great way to dock things like handheld devices, cell phones, and cameras. It would also be great for portable media. It seems like it could be called "electrical connections for dummies." I can't understand how a person could have trouble with a USB plug, but people still do. I think anyone could handle putting the camera down on the big blue pad!
Imagine a beowulf cluster. Just try not to think about replacing a brick at the bottom of the stack...
tasks(723) drafts(105) languages(484) examples(29106)
Imagine a beo....
ahh forget it you know the rest
But as a people are mentioning, what about maintenance. You have a big stack of cubes, with something wrong in the middle, you have to dissassemble a bunch to get at it. And even if the data is mirrored on another brick, what happens when you have to remove that brick to get at the dead one.
Seems to me the most efficient manner is a two dimensional spread, i.e. cover one wall with cubes to a depth of only one cube. But then in that case, you migh as well go with a traditional rack server.
Cubes are stupid especially because (think about this) even if you have a wall of cubes stacked up, if you remove one, you may have to remove all of them on top of eachother (how are these things affixed?) Wouldn't it be smarter to have it be hexagonal? Removing one wouldn't collapse a column. There are more networking connections too.
Seems like a gimmick to me. Play tetris with your SAN!
I think the issue of tripping over wires is moot. Because it's mainly for professionals in a professional environment. Initially, anyway. This is like complaining of the danger of nailguns, because most people can't operate a hammer correctly without hurting themselves.
Please stop stalking me, bro.
This reminds me of the old Gerry Anderson puppet-based TV show Terrahawks. The bad guys had evil cube shaped robots, that could stack on each other to make big laser cannons, and the like.
:)
When our data bricks start to kill people, we'll know IBM's up to something
They also talk about water cooling this system. Those connections are even harder to deal with. Hoses are always going to be thicker and more difficult to handle and there's the possibility of leaks, especially when connecting and disconnecting hoses.
Back in the 1980's there was a computer system that ran an operating system called BTOS/CTOS. The hardware consisted of boxes that could be connected together via a backplane. You could add peripherals, memory, etc, until the assemblage stretched across your entire desk.
"To be really effective, this would need to be teamed up with some sort of a backplane, but the tech is new and neat."
/. reader Shok knows better than the IBM engineers working on this technology and is paid the big bucks.
Yes, and
...These in a rubiks cube! Move them around to find the best connections incase important bricks' connections fail :P
printf("Goodbye cruel world!\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b");
Consult Rubik - LLRLR gives you financial data; LRLRLLRLR is payroll; tables are normalized if all their bricks show the same color data.
Someone else already mentioned that the article does discuss this... But I think what would look really cool in a corporate datacenter would be memory "cartridges" (for lack of a better word) like those seen in the memory bay of 2001: A Space Odyssey's HAL 9000 computer. Glass-looking cartriges that can be inserted and removed at will. You could have rows of walls with these cartridges on both sides. One can be removed and another installed, and the RAID-like setup will automatically rebuild the data at RAM-like speeds.
Others have already transferred data on notes tied to bricks -- and sometimes rocks -- thrown through windows. ...and I've frequently wanted to through a brick at Windows.
lol, so true... I think I'll just smoke reefers and play Doom 3.
Tried AO yesterday, man what a timesink.
Shame about Kieth Richards, really.
No mention is made of what happens when a brick in the middle of the cube needs
Maybe a solution to something like this would be to have some kind of rubik's cube like configuration where the data would still be accessable as long as it was connected to at least one other block, but you could move the blocks around in a preset way along "rails"
Each brick of the planned prototype will contain 12 disks and up to 80GB of storage.
Either they are using 6.6 gig drives or they are using 12, 80 gig drives....which do you think it is?
I know that they are evil, because I've seen what they do.
...the answer is: It's all automatic. The bricks rearrange themselves in mid-air, and the broken bricks fall out.
Just watch Laputa. Near the end of the movie,- you see that Laputa is composed of these very same intelligent brick computers.
In answer to the question: "How do you replace the broken bricks in the middle?"
It's true.
Just watch the movie; It explains everything.
Ice Cube brand storage bricks? A rap star's patented way of stacking weed.
God spoke to me.
"No mention is made of what happens when a brick in the middle of the cube needs to be replaced and the whole thing needs to be disassembled."
Well I got the "trite" label stuck on my response. But biological systems solve the problem with redundancy (at several levels), self-repair, and self-correction.
Plus to add. Remember in an enterprise system the worst that'll happen is that the business will fail, and have to try again later. When a biological system fails. The worst is that it dies, and there's no "do over".
Also a lot of our "new fangled technology" has principles, already discovered by biological systems, and sometimes they do them better. So maybe we need to lose the "trite" POV, and look a bit harder at what something that's been around for millions of years have to teach us.
He is also one of the architects of SPARC.
I can't seem to find the answer to this, so maybe you can help me. Do these cubes still contain disks or are they solid state?
Nowadays many data sources must run 24/7 because they serve data worldwide. It really shouldn't be necessary to shut down the data source to replace a disk. RAID on SCSI can already support hotswapping. The only nuance here is to make sure the physical design of the array allows the operator to replace any cube without removing others.
I remember reading about this here, several months ago...
Many years ago, there was a toy electronics kit like this. Each component was sealed in a little plastic brick, and the bricks were connected by magnetized plates. You could put together an AM radio, and it looked like a Scrabble board.
if these things are hot swapable, you can play towers of hanoi, and replace a damaged block when it is on the top of a tower.
It clearly shows the cooling system, which is made of sealed aluminum columns that the bricks are slid over. There are no hoses and no connections to spring a leak.
Lasers Controlled Games!
Boss: Now, Mike!!** i have always been clear tetris and other games are banned in this facility.
Mike: Boss, am trying to find the optimal block pattern arrangment that would maximise storage.
Soon, you will have a good excuse excuse for playing Enigma, Sokoban, or other box shuffling clones: "I am just excercising a failure recovery procedure for the cube in the middle, sir."
You can even play a breakout with a baseball in the real world with that technology.
There you are, staring at me again.
I wonder how long it will take for us to get to a holographic (graceful degradation of bytes if a bit goes bad,) type storage.
Instant (well light speed through a solid critalline medium) retrieval would be just what the newest DB2 database would need.
I smell copyright agreements in the wind.
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
I KNEW this sounded familiar!
See here.
From this current story, it sounds like they've made some improvements, but the two are basically the same...
bork bork bork!
I don't expect reporters (at eweek or elsewhere) to accurately report on technical details, but it would be nice if they could at least get names of people spelled correctly.
Actually, what if you had a bunch of stacks like libraries do, then to get to any cube you could just crank all of the other stacks to one side.
Those who know, do not speak. Those who speak, do not know. ~Lao Tzu
"Square bricks". Again, number-of-dimensions problem. I would think they would be cubes. Wow...80 whole gig, huh? Thanks, IBM, former leading hard drive manufacturer in an era of 250 gig hard drives! Double wow. At 80GB max per cube, you're going to need a warehouse to hold a petabyte. (13,000+ cubes, anyone?) All the benefits of plugs, but without that pesky stability of mechanical connection! Bump one with your elbow? Your elbow hurts less, and the cube goes crashing to the floor. Uh-huh. So much for that ease of installation and maintenance and low cost and whatnot.
"A great democracy must be progressive or it will soon cease to be a great democracy." --Theodore Roosevelt
ie. "To replace brick 1234, please remove brick 2345 first. Then remove 4532. Then remove 9786. Then remove 4575. Now remove 1234, replace. Now reinsert previously removed bricks"
Do not deviate from the prescribed route, or data and/or user termination may result. Please see the included reference DVD for more information about navigating to broken nodes.
Alien robots are going to come to our planet and find a way to convert this technology into a means of storing energy. Then they're going to take all of our energy, store it in these cubes, and take them back to Cybertron. Be very afraid.
All IBM has to do to trump you is create desktop sized blocks with 2TB storage per block. Given that the system needs no video system and much simpler motherboard, they can either make the cubes smaller or store more, so on a bits to bits comparison a 2TB cube is smaller than a 2TB desktop.
So to reiterate my post: Which is the simpler solution, create an array of 2TB cubes or use your solution?
GPL Deconstructed
From Jan 2002,a utonomi c_storage/CIB_Hardware/IceCubePrez.pdf
l ego_with_ice/
http://www.almaden.ibm.com/StorageSystems/
April 2002, http://www.eetimes.com/at/news/OEG20020423S0091
Jan 2003, http://www.theregister.co.uk/2003/01/03/ibm_goes_
Disclosure: I work at IBM Research where this work is done. A couple of corrections to the rather amusing thread here at shlashdot. The name of my colleague is Robert Garner, not Gardner. Each brick holds just a little bit less than a cool Terabyte, not 80 Gigabytes, i.e. the entire prototype 3x3x3 cube stores 26 Terabytes. Think of the text and the images of all the books in the library of Congress or about 6000 Hollywood movies as about that amount of data. A Petabyte (= 1 million Gigabyte) storage system, using state-of-the art disks combined with IceCube system technology is surprisingly small; I leave it to the readers to do the math. When one or several bricks fail - YOU LEAVE THEM IN THE CUBE until the machine has reached the end of its economical lifetime. Again, YOU LEAVE DEAD BRICKS IN THE CUBE. This has many benefits, which any experienced systems engineer can attest to. Of course, here is where the (software) magic comes into the picture, plus some hardnosed engineering and systems anlysis. It is way outside the scope of this thread to describe in any detail. As one british reporter once observed: 'THE ICECUBE TENET IS TO LIVE AND LET DIE'. He got it right.
...or something very similar a few weeks before they spun off/outsourced/sold their HDD manufacturing to Hitachi? The "brick stacked storage" concept thing? I seem to remember it, as I remember writing a prediction (on another forum,and I was right) that they probably had something better than conventional hard drives coming down the pike..hmmm better RTA... yep, it's an update of an older story. Better post AC then...ha
A one-dimensional wall of bricks - you mean a line?
-Mod how you like, we'll make more
It's not such a stupid idea. There are at least three applications for this technology:
Server clusters
Render farms
Storage arrays
Benefits: Zero (essentially) maintenance. All you do if you need more capacity is plug in another 'cube'. If the cube breaks, you send it back to IBM or whatever and they fix it and return it to you. It handles all the balancing and communications, and that's it.
From the outside it's like memory sticks in a motherboard; plug in a new stick and increase your ram.
Here it's plug in a new cube and increase your capacity.
Is that such a stupid idea?
What do you think is so stupid? If an idea is stupid that means you must imagine there is a better way to implement it.
GPL Deconstructed
Taking stuff apart since 1969 (TM)
I think the point is simplicity in management:
More info provided by IBM.
No wire management
No network management
No device management
It should be no more difficult to administer than a pile of Lego bricks.
I think with this design they've accomplished this.
Read up more about it. I think all your issues are addressed and moot.
GPL Deconstructed
From the photo, it looks like they're using Opto22 I/O modules as part of the system. I didn't think I'd ever see those things used in a computer setup! It looks like a fun project.
http://shit.slashdot.org/article.pl?sid=04/12/17/1 546228