Bitcoin Forum
May 08, 2024, 06:48:06 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1] 2 »  All
  Print  
Author Topic: BitIndex - The time, money, and computing power wasting database project  (Read 3559 times)
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 02, 2014, 09:20:31 PM
 #1

After a few months of inner discussion, we have decided to make it public, or i guess semi-public by starting this free blog just to update how much time, money, and computing power we have wasted just because we can, and it gives us something to do and laugh about.

For those of you too busy to click and waste 10 minutes of your time on the other blog site i will copy it below so you can waste 10 minutes here instead. ( trust when i say its a waste, unless you are looking to just kill time its not worth the read as you will gain nothing from it )

The Bit Index project is a collective of a few people who met at a bitcoin conference who found out that we all live in the same state to build one of the worlds largest and growing databases, or rather data warehouses of Bitcoin address / private key pairs in existence for ethical use only with aims of lost data recovery in the 5-10 year time frame.

The Bit Index currently utilizes the 45 drives storinator as its scalable backbone and infrastructure with each database node housing 270 Terra bytes (45 Seagate STBD6000100 6 TB drives) of  data or roughly 265 trillion address / base 58 private key pairs. With our current database design, hardware configuration and multi-threaded software interface we are able to upload roughly 20 Terra bytes of data per day taking roughly 13 days per node to fill it to max capacity.

Why?

To be completely up front, this started out as a crazy weekend project between a few friends that turned into an obsession in data warehousing. Some people collect stamps, coins, baseball cards, etc. We have become obsessed with collecting address / key pairs.

To what end?

There is no end in sight, Every day we are working on better methods of data compression to allow us to store more addresses per node, better software interfaces that allow us to get more work done per worker thread to increase import times with the current hardware configurations, and better queries to handle things like duplicate entries (if they ever occur) and data retrieval (searching the database for a specific address / base 58 key )

We are currently adding 1 or 2 new nodes per month as our time allows us to build them, right now our focus is not so much on building new nodes as it is on getting the most use per node as we possibly can. Currently we could build nodes 10X faster than we could fill them, so we are working towards better utilization before we worry about increase in current data stored.

F.A.Q.

Q: Isnt that expensive?
A: Yes, very. 3 of the 5 of us involved were very early adopters of bitcoin so a job and income became irrelevant this past year. In the scheme of things each node only costs us roughly $14,800 to build and roughly $380 a month to maintain and keep operational.

Q: What is the point of all of this?
A: For now, FUN. In the long run we aim to secure the largest searchable database of address/ base 58 key pairs known to man. Our current goal is to assemble a method of database compression that allows us to increase our storage per node to 500 trillion to 1 quadrillion addresses per node. In the end game, with proof of ownership we hope to maybe some day help someone recover their lost address on the 0.000000000000000000000000000003242% chance that we might have it in the database.

Q: What do you mean proof of ownership?
A: That is still in discussion, right now and within the next 5 years the chances of it happening are slim to none or to be exact a 0.000000000000000000000000000003242% chance at our current rate of growth that we could have your address in our database. Once it becomes a more real (over 1% chance) we will think more on how to prove ownership.

Q: Doesn't that mean that you could steal someones bitcoins?
A: Theoretically yes, but the odds are very slim, we do have a way of monitoring addresses in the database and in the past 2 months we have only seen 1 positive balance come through, of which was positive balances for less than 1 hour which lead us to believe they were part of a mixing or tumbling service sequence. No we didnt steal it, and never would. we have enough bitcoin that we dont need to. This is for fun, not for malicious purposes.

Q: Your question hear.
A: The number "hippopotamus" (yes we know hippopotamus is not a number)


Summary:

This is all for fun and just a time wasting project, a few hundred thousand dollars in cost over a few years split by 5 guys who really wont miss it much, but at the same time it keeps us occupied for a few years. We have no disillusion of "breaking bitcoin" or stealing anyone's anything, most of this is an exercise in data compression technology, and address generating software.

If you would like to contribute to the project with something useful or useless, let us know.

Ways you can contribute;

We are always looking for data to index, if you have a copy of vanitygen you could run it for an hour and make us a .txt file we can parse and upload into the database.

If you are fluent in Java or SQL, we are always looking for an extra brain to bounce ideas off of in terms of software, speed, indexing, and other what nots.

Thats about it for now.

------------------------------------------------------------------------------------------------------

Todays blog post with current progress to date (copied from the blog as well )

It's done, after 4 months of work we have finished filling our 5th node full of address / key data, and we are starting to build the 6th node this week.

As our first public announcement of this time wasting, money draining, good time having, laughable project this is more of a current status post than an update. If you read a little in the about page you will understand more of what this is and what the stats below really mean.


Current Address: 1,325,000,000,000,000
Current Storage Space: 1,350 Terra Bytes
Current # of Hard Drives: 225 Seagate STBD6000100 6 TB drives (and 10 250 Gig SSD's for redundant OS drives)
Current Chassis: Custom built clone of the 45 drives Storinator
% of all addresses stored: 0.00000000000000000000000000000009066% (rough estimate, our calculator does not have enough space to get the actual figure)

This weeks hard learned lesson: DONT TOUCH THE RED WIRE !!!!!! OUCH !!!!!!!

------------------------------------------------------------------------------------------------------

For those who read the post and missed it, we jut want to be clear, the purpose of this project is really nothing other than wasting time and money but having fun while doing it. We have no delusions that any of what we are doing will ever be useful in any way and your trolling is truly wasted on us, well maybe not we always enjoy a good laugh so if you have something original please feel free to exercise your brightest attempts at stupidity below.

Thanks for letting us waste your time Smiley we hope you got as much of a laugh out of reading it as we get out of doing it. We will keep you updated to our nonsense and wastefulness as time progresses if you care to check back at a later date.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
Fuish
Newbie
*
Offline Offline

Activity: 34
Merit: 0


View Profile
September 02, 2014, 09:28:46 PM
 #2

... Wow.
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 03, 2014, 11:23:13 AM
 #3

Had a few messages asking how we are running the address generation at current this morning when i logged in, rather than send back 6 replies i figured i would just post it.

1: download Vanitygen
2: put it in its own folder on your desktop
3: open a notepad
4: create a batch file
5: run the batch file
6: Done

To make the batch file, enter the following command on a single line of the notepad

vanitygen64 -k -o output.txt 1

then when you go to save it, switch where it says text file to "all files" and when you name it, give it a name like "Run.bat" when you click the file after saving it it will begin to generate a .txt file filled with addresses.

We wrote a java program to parse the output file into a usable .csv file that we can import into the SQL EE Database.


For those of you who messaged about donating some output files from vanitygen. Please upload them to a file sharing site somewhere and message us a link to download them. Max file size is 4GB per file. It must be the standard CPU output format not the GPU format that outputs pairs.

For those of you who messaged about helping with software and development. The database compression is kind of our pet project, things we could use is a better address generator, which is on our list of to-do's but not a priority as the main focus is first the compression in the DB. If you were talking about helping with the compression methods and schema's please message us a skype name and we will get in touch.

For the few of you who messaged us inquiring if you could pay to have access to the database. THE ANSWER IS NO. We are doing this as a project to have fun, kill time, and learn a little about data warehousing. We have not, and would not ever steal someones bitcoins just because we could even though the chances are astronomically small. If you are willing to pay to have access we are certain that your intentions are malicious which we will never support. Our location, access to anything on the same network as the SQL servers, our e-mails, pretty much everything that could lead to any misuse of the database or any theft or attacks is not and will never be allowed. Take your scamming, thieving, hair brained ideas elsewhere or go build your own database. The next time you message us with your insane ideas we will be sure and notify gmaxwell and the others with reputation enough to destroy your forum trust ratings. WE WILL NOT SUPPORT NOR ALLOW MISUSE OF OUR PROJECT.
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 04, 2014, 01:37:34 PM
 #4

*****UPDATE*****

The 6th node is fully assembled and started receiving address / key pairs as of about an hour ago, we have made a minor adjustment in the way the software layer threads the data to the drives hoping to decrease the time to fill a full 270 TB's from 13.5 days down to 11 allowing us to add and fill just shy of 3 new nodes per month.

Sometime before the week is over we will be beta testing on a 36TB mini test node, a new SQL compression algo in hopes of being able to obtain 5% more address / key data per node. This would amount to almost 210 trillion more addresses on the nodes we have built and running now.
bounst
Full Member
***
Offline Offline

Activity: 228
Merit: 100


View Profile
September 05, 2014, 08:12:43 AM
 #5

just like Coindesk bpi??
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 05, 2014, 03:00:11 PM
 #6

just like Coindesk bpi??

Wow.

 It has nothing to do with price. Its about indexing a database of bitcoin addresses and their private keys to make them searchable
Fuish
Newbie
*
Offline Offline

Activity: 34
Merit: 0


View Profile
September 05, 2014, 05:27:38 PM
 #7

I thought a large part of the draw to bitcoin was that it was secure and anonymous.... If "private" keys and addresses are searchable, even a small fraction of them, that sounds like a security and privacy hole.
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 05, 2014, 08:14:55 PM
 #8

I thought a large part of the draw to bitcoin was that it was secure and anonymous.... If "private" keys and addresses are searchable, even a small fraction of them, that sounds like a security and privacy hole.

Maybe you missed the part that even if we kept expanding this project for the next 5 years straight, the chances that we would have your bitcoin address in our database are LESS THAN 0.000000000000000000000000000003242% .

So it is very secure and anonymous to a point. Its secure in that current technology does not permit nor allow the brute forcing of a single particular address statistically. But at the same time anyone who gambles knows that statistics and luck are two different things and the reality of the matter is that even though the chances are so astronomically small it is safe to say in most cases that its impossible. It does not mean it is truly impossible for someone to generate a private key that contains bitcoin because even a 0.00000000000000000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000000000000000000000000000001% chance means that it could happen however unlikely.

I really hope you did a little more homework in bitcoin before buying into it than to not know that.
Fuish
Newbie
*
Offline Offline

Activity: 34
Merit: 0


View Profile
September 07, 2014, 12:19:41 AM
 #9

..Rude...
I am not saying it is possible or probable, I am just questioning the impetus behind setting up the operation. I also do not gamble. If I had a significant store of bitcoins, I would not want even an astronomically small chance that my keys are stored with you. Looking at if in a larger context, if someone were to gain access to this database with malicious intent, the odds of finding ANY key pair that has coins goes up greatly.
You also say "current technology", which is always changing. Who knows what future key-generation may be possible?
Just saying.
williamj2543
Hero Member
*****
Offline Offline

Activity: 588
Merit: 500

Get ready for PrimeDice Sig Campaign!


View Profile WWW
September 07, 2014, 12:21:45 AM
 #10

Just realised that with an extremely powerful computer anyones keys can technically be found eventually. Kinda scary, your coins could be randomly moved one day

█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████
▓▓▓▓▓  BIT-X.comvvvvvvvvvvvvvvi
→ CREATE ACCOUNT 
▓▓▓▓▓
█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████
mokahless
Sr. Member
****
Offline Offline

Activity: 471
Merit: 256



View Profile
September 07, 2014, 03:23:40 AM
Last edit: September 08, 2014, 02:38:47 AM by mokahless
 #11

Terra is land or Earth, based on the French word for Earth
Tera is an order of magnitude.

Might want to fix that. On your site and this forum. Doesn't reflect well on the team.

Quote

We're not here to impress or stun with our cunning English. But if you are applying to be an editor free of charge for all future posts you are more than welcome to it. Smiley

There are a whole bunch of other English issues on your site and post but no one, including me cares. Terra and tera- is a pretty glaring mix-up though, especially when it is in big green title letters on your site. I don't want to be your editor, I just wanted to point out an error. You're welcome. I can see you take criticism well by sarcastically passive-aggressively attacking those that comment.

BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 07, 2014, 12:31:49 PM
 #12

Terra is land or Earth, based on the French word for Earth
Tera is an order of magnitude.

Might want to fix that. On your site and this forum. Doesn't reflect well on the team.

We're not here to impress or stun with our cunning English. But if you are applying to be an editor free of charge for all future posts you are more than welcome to it. Smiley
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 07, 2014, 07:16:19 PM
 #13

..Rude...
I am not saying it is possible or probable, I am just questioning the impetus behind setting up the operation. I also do not gamble. If I had a significant store of bitcoins, I would not want even an astronomically small chance that my keys are stored with you. Looking at if in a larger context, if someone were to gain access to this database with malicious intent, the odds of finding ANY key pair that has coins goes up greatly.
You also say "current technology", which is always changing. Who knows what future key-generation may be possible?
Just saying.

It is a really expensive experiment and nothing more, we too understand and had the same concerns with the database project as you do and that is why it is completely cut off from any network or internet access so it could never be hacked into as it is not, has never been, and never will be online. Furthermore the only people with access to it are the ones building it and we have no need to steal anything from it. Although that is not to say if the address for the DPR coins or something along those lines came up in the DB that we would not take them back and distribute them back to the BTC community as we believe that the fashion in which the US marshals have found to sell them before guilty verdicts are placed is criminal in itself.

Future key generation is not far off, one of our team who is not fluent in DB languages is currently working on an FPGA solution to generating addresses in the long term as our storage solution problems are fixed. It should allow us to build a new node and generate enough addresses to fill it at a rate of 1 node per day, although for that to happen we would also have to solve the .csv import to SQL EE problem as even with 100 threads on 4 12 core processors and 256 ram it is still a very long and painful process to get the files we generate input into the DB.

Even still a series of FPGA address/key generators still would not be really threatening to BTC as we are still confined by the limits of a computer as the FPGA's will only handle the IO and computing and the software interface still has to run multiple threads writing multiple files to keep up with the rate that the FPGA is cranking them out and you can only get even the biggest server processors to write those files to ssd's so fast.

On an insanity scale, even spending millions of dollars on developing an ASIC which does nothing but run the IO for computing addresses, the time to generate a specific individual address even with 50 TH/s of address generating ASIC power would be something to the tune of a few hundred million years.

All in all even within the next 5 years if we keep up at 300% of what we can produce and store now. Its really nothing to be concerned about, like we said, we got lucky, or whatever you want to call it and got to witness one of the addresses in our database receive then about an hour later send bitcoin to a different address. That was pure luck and not likely to ever happen again as it was never likely in the first place.

Furthermore, the amount of people who have the amount of money something like this requires are limited. Usually scammers and schemers are out to make a buck because they are too broke to do anything else productive so they could not afford the hardware to do something like this on the next to nothing chance they ever see a penny of what they spent back. So that leaves it to guys like us who have a lot of money that pretty much was all pure luck in itself when bitcoin became worth something real, that are more curious than we are spend conscious and dreamed up an insane, fun, laughable project to keep us busy for years to come.
ajmssc
Newbie
*
Offline Offline

Activity: 1
Merit: 0


View Profile
September 08, 2014, 08:55:34 AM
 #14

If your database is offline then how did you detect the address match in the chain? And if your purpose was truly not for profit why would you even scan it?
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 08, 2014, 03:37:52 PM
 #15

If your database is offline then how did you detect the address match in the chain? And if your purpose was truly not for profit why would you even scan it?

Part 1: If your database is offline then how did you detect the address match in the chain?

The database is hard wired to a node which is wired to a relay which is wired to a machine that runs the insight API which runs from a remote database which is populated by a remote bitcoin client connection.

Part 2: And if your purpose was truly not for profit why would you even scan it?

Research and historical data collection.
williamj2543
Hero Member
*****
Offline Offline

Activity: 588
Merit: 500

Get ready for PrimeDice Sig Campaign!


View Profile WWW
September 09, 2014, 12:45:45 PM
 #16

"We will use this for password recovery". Sounds like your trying to brute force addresses for profit?

█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████
▓▓▓▓▓  BIT-X.comvvvvvvvvvvvvvvi
→ CREATE ACCOUNT 
▓▓▓▓▓
█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 09, 2014, 11:18:17 PM
 #17

"We will use this for password recovery". Sounds like your trying to brute force addresses for profit?

Sure why not, we were just super slow to the draw last time when one of the addresses in the database had a positive balance and totally forgot to steal the bitcoins? You didn't read much did you?

Let me recap for you:

-----------------------------------------------------------------

Maybe you missed the part that even if we kept expanding this project for the next 5 years straight, the chances that we would have your bitcoin address in our database are LESS THAN 0.000000000000000000000000000003242% .

-----------------------------------------------------------------

For the few of you who messaged us inquiring if you could pay to have access to the database. THE ANSWER IS NO. We are doing this as a project to have fun, kill time, and learn a little about data warehousing. We have not, and would not ever steal someones bitcoins just because we could even though the chances are astronomically small.

-----------------------------------------------------------------

Q: Doesn't that mean that you could steal someones bitcoins?
A: Theoretically yes, but the odds are very slim, we do have a way of monitoring addresses in the database and in the past 2 months we have only seen 1 positive balance come through, of which was positive balances for less than 1 hour which lead us to believe they were part of a mixing or tumbling service sequence. No we didnt steal it, and never would. we have enough bitcoin that we dont need to. This is for fun, not for malicious purposes.

----------------------------------------------------------------


If any of that from above was not clear enough, we are not brute forcing peoples addresses in hopes of stealing bitcoin for profit. As for account recovery, it is a far cry from anything we could accomplish right now with only 6 nodes and only 5.5 of them full so far. Honestly in a hundred years the chances that we have the key to someones address who actually lost it is slim to none.

To clarify and make it a little more clear. None of this is for profit, we could find better ways to spend hundreds of thousands of dollars trying to make an ROI then cataloging bitcoin private keys. Hell just hanging on to the bitcoin and waiting for the price to rise again would be 10 million times more ROI or profit than this project would ever make us.

BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 12, 2014, 11:08:43 PM
 #18

**** UPDATE ****

The 6th node is almost halfway filled with just under 134TB stored already. About 1 more week until it is fully loaded with addresses.

In other great news, we have just ordered all the components for the first of the new 360TB nodes which will house 60 WD 6TB drives each. We are expecting the components to arrive next week so we can begin assembly.

and finally, We have decided we are going to re-name the project as many of the community here who have jumped on board with helping on the project in different ways have said it needs a better title. We should have one by the time of the next update.
tucenaber
Sr. Member
****
Offline Offline

Activity: 337
Merit: 252


View Profile
September 13, 2014, 01:09:44 PM
 #19

/.../ historical data collection.

This doesn't really make sense does it? The only data not already in the block chain is random data you generate yourself  Roll Eyes
BitcoinCitizen86403 (OP)
Newbie
*
Offline Offline

Activity: 23
Merit: 0


View Profile
September 13, 2014, 01:40:21 PM
 #20

/.../ historical data collection.

This doesn't really make sense does it? The only data not already in the block chain is random data you generate yourself  Roll Eyes

Historical data =

# of times a birthday attack would have succeeded on the blockchain (if our intent was to steal something)
# of address collisions, & # of Address & Key collisions
# Odds vs. successful collisions data points. (showing that the math can be beaten)

Please show us where any of the above data is already in the blockchain, and for the hundreds of thousands of dollars you save us, we might just share some with you Smiley
Pages: [1] 2 »  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!