How much hashing power the CIA can organize ?

[ArsenShnurkov]

In one russian thread we need this data to discuss possible variants of future.

Timeframe - to the end of year 2011.

Bounty - 8 BTC for sound fact-based estimation.
Chinese guys - what do you think?

[TiagoTiago]

What about the FBI? Or do they share resources with the CIA?

[ArsenShnurkov]

What about the FBI?

You can do, but estimation will not be awarded by me. May be it is interesting to someone else.

[barbarousrelic]

I would think the NSA would be the agency with the most potential hashing power, since they specialize in codebreaking.

[CCCMikey]

So I guess 'the more you mine, the safer the currency.'

[Freddie]

Firstly, as barbarousrelic mentioned it is the NSA, as opposed to the CIA, who command most of the available computing power.  However, from what I can gather, both are believed -- at least in the past -- to command similar budgets [1].

So, how much computing power could they command?  Well, firstly, supercomputers are off the list.  A Top 500 supercomputer, the CX2 at Imperial College London has around 10,000 cores (Intel Core 2 era) [2].  Lets say each core can do 1mhash/s; this gives us ~10ghash/s.  Compare this to a high end GPU (one 'core') which can do 150mhash/s.  We break even at around 70 GPUs (or 45 if we use dual-GPU cards, even less if we overclock).

In other words: forget supercomputers; they're out.

So, what can take on a GPU?  Well, lets take a look at some dedicated FPGA's (field programmable gate arrays).  The NSA@Home project uses 15 Virtex-II Pro FPGA's (originally intended for video transcoding) to crack SHA1 and MD5 hahes.  The result is claimed to be equivalent to 1500 AMD FX CPUs and requires on the order of 240 W [3].

When analysing this two points are key.  Firstly, the FPGAs used are not even close to state of the art.  More modern FPGAs will not only be capable of doing an entire SHA256 round in a single cycle but will also be clocked higher.  Secondly, the board on which they are mounted was not designed for hashing -- but rather transcoding.  Now, the FPGAs are not cheap (however, I can't find a price for the specific model used in NSA@Home which appears to have been discontinued,) but they can be retooled (so are something of a long term investment for an agency).

If a large quantity are required it becomes cheaper to fab an ASIC.  The cost for such an operation is easily in the $ mln range, but is far cheaper when it comes to bulk production.

Given a reasonable five-figure sum I am confident that in 3-6 months I could have a working FPGA for bitcoin mining.  Even less if existing SHA256 IP were used.  A simple Google search for "SHA256 FPGA" will show that there is no shortage of existing work, from both academia and industry, on the subject.  It requires expertise and is far from trivial, but definitely doable.

It is therefore almost certain that if the NSA (or others) wanted to attack bitcoin they could do so.  It would probably take no more than a month to retool their existing FPGAs and bring them online.  While I will not speculate as to the theoretical hashing power I would expect it to be in the thash/s -- easy.  Such capabilities are also in reach of multinationals with expertise in the field of integrated circuits or even well funded individuals with too much time (and hardware!) on their hands.

Regards, Freddie.

[1] - http://www.fas.org/irp/commission/budget.htm (Outdated.)
[2] - http://www3.imperial.ac.uk/ict/services/teachingandresearchservices/highperformancecomputing
[3] - http://nsa.unaligned.org/index.php

[dishwara]

At least a minimum of 1-10 TERAHash/s can be easily done by CIA.

[FatherMcGruder]

So, every little bit counts right? Let's keep the CPU generation option in the standard client.

[barbarousrelic]

Does a purpose-built chip for SHA256 hashing exist? Or anything better than GPUs?

I'm wondering if someone like the NSA couldn't make purpose-built circuit boards of such chips that could blow away the GPUs commonly used now.

[dbitcoin]

It's much simpler for such organization just buy same amount of GPU's.
Bulk price for new 5790 probably 450 now.
For current network hash power  471 Ghash/s, we need less than 1000 5790 vga cards.

1000 x $450 = $450k for cards
500 x $300  = $150k for another components for dual gpu rig.
$450 + $150 = $600k
But for attack we need only more than 50% hash power:  $350k enough.
So there is no problem kill bitcoin currency.

But you really think government should cares about potential $21M niche market?
When government throws away billions for another small war?
When just printing another 600 billions for bankers?
Or spends more than 300 billions for support japan market and around 150 billions for reconstruction?
Right now bitcoin it's a kids games in government eyes, and nothing more.
And in the near future nothing change (IMHO).

[Freddie]

Does a purpose-built chip for SHA256 hashing exist? Or anything better than GPUs?

I'm wondering if someone like the NSA couldn't make purpose-built circuit boards of such chips that could blow away the GPUs commonly used now.

Yes, as my post outlined FPGAs (field programmable gate arrays) and ASICs (application specific integrated circuits) are to a GPU what a GPU is to a CPU insofar as performance/watt is concerned.  ASICs, however, are extremely expensive to tape out, require extensive debugging and testing, and are fixed function.  FPGAs -- as the name implies -- can be field programmed -- and so can adapt quite readily to new algorithms.

The chips do indeed exist -- IP (intellectual property, i.e, verilog or VHDL source code) for SHA256 cores is available from many companies -- and can be used to either program an FPGA or to fab an ASIC.

Regards, Freddie.

[Jered Kenna (TradeHill)]

It's much simpler for such organization just buy same amount of GPU's.
Bulk price for new 5790 probably 450 now.
For current network hash power  471 Ghash/s, we need less than 1000 5790 vga cards.

1000 x $450 = $450k for cards
500 x $300  = $150k for another components for dual gpu rig.
$450 + $150 = $600k
But for attack we need only more than 50% hash power:  $350k enough.
So there is no problem kill bitcoin currency.

But you really think government should cares about potential $21M niche market?
When government throws away billions for another small war?
When just printing another 600 billions for bankers?
Or spends more than 300 billions for support japan market and around 150 billions for reconstruction?
Right now bitcoin it's a kids games in government eyes, and nothing more.
And in the near future nothing change (IMHO).

It's not that they want the money. If Bitcoin is really successful then it would be a threat to the dollar and the power that it holds. Not to mention all the "evil" things it can be used for like money laundry, drugs, terrorism or whatever they want to claim.

[LMGTFY]

It's not that they want the money. If Bitcoin is really successful then it would be a threat to the dollar and the power that it holds. Not to mention all the "evil" things it can be used for like money laundry, drugs, terrorism or whatever they want to claim.

All good arguments for the state wanting bitcoin dead, and I can certainly imagine the FBI and NSA acting covertly against bitcoin, but the reasons you give seem to me like reasons why the CIA would privately like bitcoin: money laundering - ideal for paying assets; drugs - ideal for financing covert operations; terrorism - "one man's terrorist is another man's freedom fighter", etc.

I wonder... are there CIA agents using bitcoin, right now...?!

[dishwara]

So, its like states created bin laden for back ground operation & until he was working for them, he is gentleman. But when he was against state, he became terrorist.

Similarly, until bitcoin helps state(CIA, FBI, ABC, XYZ), they will let it grow, once it threatens state then they will act against bitcoin...

[ShadowOfHarbringer]

If Bitcoin is really successful then it would be a threat to the dollar and the power that it holds.

By the time Bitcoin becomes "really successful", the network will operate at a hashing power of at least 10-100 Terahashes / sec and it will keep growing quickly.
It will be impossible to destroy the network then.

Another thing is that if CIA/NSA/FBI/Whatever will keep constantly attacking Bitcoin, then we will surely develop new algorithms to counter that. The only reason we aren't doing it yet is that there is no need for it.

[Anonymous]

How long till a politician gets caught mining bitcoin lol

[gigabytecoin]

With a $10,000,000 budget dedicated to buying 5970s, one could easily create a 10 Tera Hash per second computer cluster that would immobilize the bitcoin network.

See my original post here: http://bitcointalk.org/index.php?topic=4559.msg67079#msg67079

[Anonymous]

With a $10,000,000 budget dedicated to buying 5970s, one could easily create a 10 Tera Hash per second computer cluster that would immobilize the bitcoin network.

See my original post here: http://bitcointalk.org/index.php?topic=4559.msg67079#msg67079

So that means the bitcoin network would be sitting idle. What if all those idle computers were pointed at the attackers connection ?

cough*ddos*cough

Maybe instead of being frightened into submission how would a defensive action be taken ?

[marcus_of_augustus]

I personally would not be surprised at all if Satoshi is a front for some 3 letter agency and bitcoin is a plot to get a covert way of financing things for them. Use of sha256(sha256 hash is telling too.

Now you're getting somewhere .... there are competing interests and factions within the "agencies" and "the company" itself.

What if one faction actually believed that fiat-debt currency was ruining the world (which it obviously is) and set bitcoin loose as competition to the dark-side factions who are desperately trying to cling onto control through the status quo, murder, corruption, manufactured crises, etc ...?

[Anonymous]

The internet itself was started by the military....look how that turned out.


Beauracrats are too incompetent to install bitcoin and would probably do it wrong anyway.