Large Scale Solid State Memory Storage? 76
spacechicken asks: "I am doing a theoretical study of an extremely secure large scale data storage concept. Due to the nature of the (theoretical) location of the (theoretical) warehouse some of our constraints include very few (if any) service visits, complete remote administration and no moving parts. Does anyone have any experience using or information on large scale (on the order of 10^12 - 10^15 bytes) deployment of solid state storage? And to preempt those who will say it - I have Googled."
Here you are. (Score:2)
The technology is quite mature.
Re:Here you are. (Score:2)
HUH?!?! (Score:2)
That's about a quarter of a billion dollars in PC133 ram. I'm sure wholesale prices would bring that down, but you're still talking many millions of dollars.
Heh (Score:1)
Re:Heh (Score:2)
The biggest consumer drive I know of is Maxtor's 320GB drive. It would take a minimum of roughly 3000 drives to cover ~900TB with no redundancy. The MTTF of these drives is over a million hours according to Maxtor, though that's over a hundred years so I question this figure. I wasn't able to find a price on the 320GB drives, but assuming double the cost of the 160GB drives, we're talking 3.2 million dollars for 2x redundancy, much lower than the 250 million price for single-redundancy in memory.
As you can see, the cost is still enormous even with non solid-state. Maybe the article submitter needs to delete a few (theoretical) things so he doesn't need so much (theoretical) space
Re:Heh (Score:2)
Re:Heh (Score:1)
here is [pcguide.com] a good article on MTBF
Re:Heh (Score:1, Insightful)
Re:HUH?!?! (Score:2)
The guy who submitted this story is talking about over 900000 gigabytes of storage, using only memory and no hard-drives. Using modern technology, such a thing would cost shitloads of money. I came up with the figures of over 3 million dollars to do it with hard-drives with 2x redundancy, and half a billion to do it with PC133 memory with 2x redundancy.
Re:HUH?!?! (Score:2, Informative)
Never mind the cost. Right now we are looking at the technical problems.
As I said - there will be no visits from service staff to replace/repair failed modules. Also, the site will need to be set up in one go - hence the need for so much storage all at once.
Hard drives can not be used - they would not survive the trip or the location.
And I am most definitely NOT Uncle Sam.
Re:HUH?!?! (Score:2)
Sounds like a space/deep ocean shot of some sort. Curious, curious... Let us know what it is?
Re:HUH?!?! (Score:2)
Sounds like a space gig, based on the poster's previous post in the Australians to Build Spaceport on Christmas Island [slashdot.org] thread. Also, in this thread [slashdot.org], the poster specifically uses the example of Australian Aerospace students.
While circumstantial, I would guess that this is a privately funded space shot. Why he needs this much storage is still an open question, but by non-serviceable I believe we can take him literally
Re:HUH?!?! (Score:2)
Re:HUH?!?! (Score:2)
Re:HUH?!?! (Score:1)
The ultimate remote backup site. After all if this 'site' gets blown up, at least your "data" collection will be safe!
A couple of things if you are:
You will need to make sure all the stuff is radiation hardned. Not so much for when it gets there (you can bury it), but for the journey out.
If you bury it, then you will have serious heat problems.
Solar power is a reasonable start, but don't forget the cells will degrade due to the solar wind and micrometeors.
Once you have all the equipment set up, how are you going to maintain a connection to earth? You will have pretty limited options, line of site laser to sats, then some sort of sat up link down link, not terribly practical for >10TB data store you suggest.
If you're not thinking of a lunar data store then ignore all of this and wipe it from your memory.
Re:HUH?!?! (Score:1)
Should not be too difficult technically. i'd build it in the form of :
dual CPU mobo with 1 terabyte of memory with 8 cards of 10Gbps fibre channel (to survive radiation/EMP) networking gear. x 1024
assuming minimal boot from a solid state flash drive and the terabyte of memory can be addressed as a linear block if the CPUs are 64 bit, the dual procs should give enough capacity to saturate a 80Gbps (10gigs/sec) uplink. connect em all in the form of a hypercube and ensure that you have at least 8 different routers connected to each port of a single node. if any memory modules die, the ECC on the mobo (and chipkill type tech if you have it) should take care of it. If the CPUs die you loose 1 TB of storage. If 1-2 network adapters die any decent OS (like Linux) can isolate em. Since youre using at least 8 different routers PER mobo, a router dieing will reduce interconnect bandwidth (by 10Gbps) but wont isolate any nodes.
Multiply this entire setup by 3 if you want extreme redundancy and mirror the data 3 times on 3 different hypercubes.
Not a real problem. Mobos should be cheap (use MIPS or similar low powered/low heat 64 bit CPUs), memory reasonably expensive, 10Gbps ethernet cards + routers cheap, a small UPS+PSU per mobo should also be sufficiently cheap (you get integrated UPS+PSUs now). If you want this setup to survive power failures simply add 1 TB flash per node and write to it if the UPS throws an alarm. Controller software for the hypercube is easy to write if you know what your doing..i did something similar which can be adapted for exactly this setup a while ago.
Re:HUH?!?! (Score:2)
Re:HUH?!?! (Score:1)
If you go at the other side of his range, it's only 930GB. Knowing that I bought 256MB PC133 RAM for $CAN32 last week, that makes it ~$CAN120,000, or ~$US75,000. And that price is street price, not wholesale.
Now, of course you'll need backplanes, electricity, etc., so the cost will be higher. But with a range of 3 orders of magnitude, it's normal to have a price 3 orders of magnitude bigger.
Back to the topic... With such a big deployment, you're bound to have some failures every few days. Do you have a way to replace the defective parts, or since there's no moving parts and no human intervention it's not possible?
Re:HUH?!?! (Score:1)
I'd suggest you also add lots of impact shielding. Even if you put up a bunch of replicas, over time, they're all going to get little holes poked in them.
The alleged "joke" posted earlier (probably modded to oblivion... I don't know, i surf noscore, -1, hidemods) about clay tablets bore a grain of truth. Large area/byte, physically robust, soft enough to not shatter on impact from extremely small, high-velocity projectiles, large reading interface tolerances.
Should be easy enough... (Score:3, Funny)
Hope that helps.
Geez. (Score:1)
*pulls out a calculator* Hmm, 10**12 bytes is approximately 931 TB.
WHATEVER FOR?!
And that's on the low end, too. Since you're ranging a couple orders of magnitude, you might be pushing the limits of recognized abbreviations. All in solid state, too? That could be costly.
Re:Geez. (Score:1)
Re:Geez. (Score:2)
Re:Geez. (Score:1)
Re:Geez. (Score:2)
Re:Geez. (Score:2)
Ummmm hello?!? alt.binaries
How about RAMDRIVE + SCSI + SAN? (Score:2)
For example
http://www.storagesearch.com/ssd-3.html can hold up to 8gb per unit, and at 15 per SCSI card, 6 cards per system, you'll end up with 720GB per 'server'.
Figure gigabit fiber (at the minimum) between machines and you'll end up with quite a bit of storage.
Re:How about RAMDRIVE + SCSI + SAN? (Score:2)
Question (Score:4, Interesting)
So I'll make a guess. 10^12--10^15 bytes is a large range. And I can only think of a few ways to generate that much data. The most probable is video cameras, but I can't think of any reason why you would need it secure in that fashion.
Secure without human intervention is interesting. I mean, if all you want is security, the easy way is distributed networks and encryption. And really, that would be more secure in the event of nuclear war or other similar events.
So I have two guesses:
A) Given the similarity of your numbers to the 10^11 neurons in the human brain (and each neuron has as many as 1000 connections to its neighbors) this is some sort of screwy immortality thought experiment.
B) Given the security requirements, this is some screwy thought experiment involving the preservation of the sum of human knowledge over a vast stretch of time without human presence. That could be interstellar travel, say, or large disasters wiping out the human race.
Re:Question (Score:2, Informative)
Rest assured it is neitherone of the two screwy thought experiments you have postulated. It is another one entirely. Albeit one with a commercial bent.
As I said earlier the reason for so much storage is that the site needs to be built in one go. There will probably be no ability to increase the size of the facility as requirements grow. We are not pulling numbers out of the air (or counting neurons). We are simply planning ahead
Re:Question (Score:1)
Re:Question (Score:2)
Re:Question (Score:1)
You said that mechanical mechanisms would survive neither the final environment nor transportation there. I can think of a few places which would stress drive mechanisms, and which would be difficult/impossible to reach for repairs. Ocean floor, Arctic or Antarctic, deep space,
But I can't think of which one of these would have a "commercial bent" (at least not with a good ROI for a one-shot deal).
<voice type="Dr. Evil">
Throw us a frickin' bone, here!
<\voice>
-Ster
Re:Question (Score:2)
Re:Question (Score:2)
Back of the envelope calculations (Score:2)
Assuming slow, but commodity CompactFlash cards of 1Gigabyte each (currently $800 retail, your price may vary). You'd need 1000 to 10^6 of these puppies for approximately 800,000 to 800,000,000 $US (Retail). It would be fairly compact, fairly reliable, and fairly slow.
So, with a price of $800,000 today for the low end, in 3 years (more or less) the price for 1 Terabyte of CompactFlash will be $100,000. This drops further to the point where I can afford it, in about 2 more orders of magnitude (7 years?).
Bottom line, it's feasable, would be $1Million to $1Billion to implement 1 of (at retail, buy the fab, prices WILL drop dramatically). I'll be able to afford the same thing 10 years later.
--Mike--
Re:Back of the envelope calculations (Score:1)
Okay, so here's what you need to do. If you can stand a little latency and have a lot of money to burn. Launch a satellite into geosynchronous orbit that's always visible from the theoretical warehouse, that just echoes the signal back (and amplifies it of course). Broadcast out the data stream at as high a bitrate and over as much of the spectrum as you can cover aimed at the satellite with the intention of the satellite bouncing the same signal(s) back to you in a little amount of time. Thereby, you can get the following amount of storage out of the system (theoretically, just hope for no planes, you might want another satellite for redundancy):
2 x total bitrate x distancebetweenstatelliteandwarehouse / speedoflight
and a maximum latency of:
2 x distancebetweenstatelliteandwarehouse / speedoflight
of course if this "warehouse" is going to be moving, then you'd probably want a pilot signal to detect and compensate for any shifting. Now how easy is that
Re:Leg work on /. leads to fires of speculation (Score:3, Informative)
There is a serious topic behind this question. However, it usually requires a lot of explanation.
Some of your guesses are correct - this does have space applications, but not in the way you think.
Suffice to say that the storage is not planned (theoretically) to be used as anything more than a commercial repository.
The reason for the vast amounts (as I keep saying) is because there will probably be only one visit to the site to install it all - meaning no upgrades.
Re:Leg work on /. leads to fires of speculation (Score:2)
Heh, screw HavenCo, let's see the feds get a search warrant for an orbital datadump.
On second thought, one would have less social protections against getting the crap nuked out of it...
Re:Leg work on /. leads to fires of speculation (Score:2)
Re:Leg work on /. leads to fires of speculation (Score:1)
There is a way to protect it from radiation though - both natural and "man-made".
Re:Leg work on /. leads to fires of speculation (Score:2)
A)You are sending it up in a big lead box.
B)You are burying it on the moon or another spacebone body.
C)If ISS has proper shielding, it's going into ISS.
A would probably be insanely expensive. Costs a lot to shoot lead into space. That's why NASA buys radiation-hardened equipment instead of buying some P4s and shooting them up in a big lead box.
B would be huge news. It would be extremely likely to fail if you send a robotic digger to the moon or an asteroid, so you'd have to do it with people, which would be a historic event.
C makes the most sense, but I don't know if ISS has enough shielding. Presumably there's enough to keep the people from overdosing on bad particles, but I'd imagine the computers would still have problems.
Tim
Re:Leg work on /. leads to fires of speculation (Score:1)
Actualy, all solid state memory experiences errors due to cosmic ray particles, against which you CAN'T shield- eventually, some of these high-energy suckers will get through- and the problem gets worse the higher you go.
The chance for a given memory to fail due to this reason is called MTBF- Mean Time between Failures (actually, there's a broader definition [techtarget.com], but I'm using the one related specifically to memory).
In addition, the more memory you have, the more errors you will have for the same MTBF- for example, if the MTBF is 1000 years for a single MB of your ultra-shielded memory. For 1000 MB, that means almost certain failure once a year! and you are talking about MUCH larger memory sizes!!
To conclude- in space, no one can hear you scream...
To Probe further:
Cosmic Rays [ualberta.ca]
An article called "Can Hardware Be Trusted" [embedded.com]
Despite everything I said above, there has been research on fault-tolerance in space, which might help you. You can look at the homepage of the Stanford REE project [stanford.edu] for more details
You might also be interested in these slides (PDF document) of a research project called Fault-Tolerant Computing for Radiation Environments [stanford.edu].
Hope this helps
Astromage
Re:Leg work on /. leads to fires of speculation (Score:1)
META: How to put links in /. posts (Score:1)
<a href="http://www.yourstuffhere.com"> Fancy Word Description</a>
You have an "Allowed HTML" reference below you, I won't go to deeply into it, the B tag is for Bold, I for Italics, P for Paragraph, try them out, but do remember that all tags delimit an area of text and so they have to have an end tag. For instance to make a word bold you do <B> this </B> and it should look like this.
Any more questions feel free to ask.
i got it! (Score:2)
Ok, what about other problems? (Score:5, Informative)
1: Power. Solid state drives tend to forget stuff when shut off, so you'll need a UPS in the data center to handle it. No biggie, except when you realize that the batteries are going to need maintence. They do go bad after a while, I know of no batteries that don't. In theory you could do flash memory instead of volitile for about 3 times the cost (512meg ATA flash storage is $300 on pricewatch, add raid, san, etc, pricey but possible)
2: Cooling. Massive amounts of solid state chips are going to generate massive amounts of heat. This means water chillers (most likely) and fans. Both involve moters. Moters go bad. You can build redundent, but eventually both cooling systems will go out.
3: The hardware it's self. CPUs go bad, controllers blow up, ram chips go out, power supplies blow, etc. You can only leave a redundant system alone for so long until it's no longer redundant.
4: Acts of god. Floods, fire, lack of fuel for power, emi. These things happen. You can build two data centers in seperate locations and write one off when something bad happens, but then you're back to no redundancy.
5: Murphy's law. Don't forget, Murphy always wins.
Having worked on a LARGE scale redundent system (Think uncle sam), I can tell you these things do require maintence. Building a system that large without bugs that creep up in a few years time is going to be next to impossible.
That said, it sounds neat. Let me know when you guys need an engineer to build it, it'd be fun.
Re:Ok, what about other problems? (Score:1)
2- Cooling - When you have power...cooling is easy..
3-4-5 - Yep...
Not much to say (Score:2)
Remotely administering solid-state storage is no different than remotely adminstering spinning storage, just less chance of failure, so less visits. The data size you require can easily fit in a single rackmounted disk array. Vendors like E-Disk sell highly reliable solid state drives in IDE and SCSI, up to at least the high 10's of gigs in size per disk, 2.5 and 3.5 in form factor.
Re:Not much to say (Score:2)
I think your calculation may be broken. Spacechicken is asking about storage in the range of one terabyte (lower bound, very easy to achieve) to one thousand terabytes (definitely possible, but bigger than damn near everything). You can squeeze about 1.8 TB into a rackmounted disk array using 160 GB drives; the new 320 GB drives will double this to about 3.6 TB, at least in theory. You're still 996.2 TB short of Spacechicken's high number.
Re:Not much to say (Score:2)
Oops, I was off by 10^3 when I read his post
Still, that just means you need several racks, and the problem against isn't solid-state specific. Check out EMC and Hitachi for better storage density in a normal array, but they may not offer a stock solid-state option (although I bet you could retrofit it with drives yourself). In any case, managing a large amount of storage, and managing solid-state storage, are two different things, both of which are pretty standard question with standard answers.
I will say one thing - if you're thinking of SANing all of this storage to a cluster of servers - don't buy the existing market leaders' products. Cisco is on the verge of coming out with their second-gen SAN switches that are in catalyst chassis and support virtualization (read: veritas software-raid) inside the switch, making for a very nice solution.
Space (Score:2)
The only other practical place I can think of right now where hard drives wouldn't survive the trip is some sort of undersea base. Sealand already has an offshore data haven, and they can visit theirs.
Tim
Look At Me! (Score:4, Funny)
Look at me Mom, I got my question posted on Slashdot by makin up some bogus and totally illogical question. Then I made tons of vauge references and quasi-logical posts to make people wonder. Ooh ooh ooh look at me!
Face the truth (Score:1)
(If it's not clear to most people reading this, the questioner wants to build a device that can listen in on a fiber optic tap and record a lot of data for a long time. Since fiber is vulnerable where it lies on the bottom of the ocean, and that position would necessitate the other requirements, that's the best guess.)
Talk to IBM (Score:2)
An actual answer (Score:2)
Anyway, at least one solution to this request appeared on Slashdot just a few days ago: The solid-state RocketDrive [cenatek.com].
Perhaps not the ideal solution (I honestly can't say if OEM solid-state storage exists on a much bigger scale), but something that you can concretely say "this would work". Proof-of-concept, if nothing else.
Granted, for the size you want, at $5k/4G, This would cost USD $1.3 billion just for the storage itself (not counting the array of 32k+ PCI slots you'd need to hold all these and the hellacious network to RAID them), but this sounds like a gub'mint project anyway, so cost presumeably forms the *last* of your concerns. If cost *does* matter, you can get the unpopulated controller boards for $800 each, and certainly a *much* better bulk deal on RAM then what Cenatek offers (basically they charge $1k/1G? Perhaps 10 years ago!).
Checking Pricewatch, the average non-volume-buyer can get 1G of PC133 for around $100). That would lower the storage-only cost to only USD $315 million, before considering volume discounts.
this whole thing is a massive troll (Score:2)
The only storage that would begin to meet SOME of the requirements would be molecular/biological. Try again in a decade.
If cost is not an object... (Score:1)