Interchangeable Data Storage Bricks? 185
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."
No mention of... (Score:5, Informative)
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.Very Tough Error Isolation (Score:5, Informative)
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.
In case the article gets slashdotted (Score:1, Informative)
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.
Seems like a solution looking for a problem. (Score:4, Informative)
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.
Re:Mr. Obvious says... (Score:3, Informative)
Actually, I'll tell you how many dollars nasa spent developing a pen that worked in zero g. They spent 0. Not a cent. Someone developed it on their own for scuba diving, and then they bought tons of them from him because pencils (which they were using) have all sorts of problems in zero g - for instance, graphite and wood shavings could get in circutry, both are flammable, etc.