Coding Well Forum

You could have clients "check in" every so often and see if the computation they are doing is worthless. Non-Primes would be easy to verify, The server could do it in like 1 second. And we could do ones we think is prime twice with different comps each time.

My First post here:

A system like this would be great btw good luck

but what to do on it - now theres a question

Ive recently been learning about Hash's - so

i know that:

[*]hashes always spit out the same number of characters in the result
[*]Hash x = Hash x
[*]theres no dehash function

and
[*]hashs can be easly brute-proccessed if you know the salt and posible combonations (i created a bank for EVE-Online - use hashs to store pin numbers 9999 combonations )

Now we know that (from number 1)
Hashes always spit out the same number of characters in a result

SOOOOOO
theres obvoisly going to be dupicate hashes somehwere

cos x amount of chacters (hex or not) has a finite amount of combonations

So
-Mission-
find duplicate hash's with .... Sha512

There are already projects for that and I don't have anywhere near the storage needed to do that (Terabytes)

And it is computationally infeasible.

Its none the less posible, finding duplicate hash's
and if you spred the load abit, over mutliple DB servers - i'd be happy to let you use mine (yeah ive got one now )

Try a smaller hash then, like Md5 or CR32

md5 is done. sha1 or 2 would be the next best. We would likely need to roll our own database solution.

OR

just use mysql - a MyISAM table should have no trobble doing this
now untill i have done the math yeah.... but if you can afford some of the space go for it

alec

You would have to figure out how it would scale.

Store it across multiple database computers, liek google , think its called shadowing

for example

send to database one: select * from hash_list where hash = $hash

send to database two: select * from hash_list where hash = $hash

send to database three: select * from hash_list where hash = $hash

send to database four: select * from hash_list where hash = $hash

fi any of these return a value of one youve foudn a duplicate
send to hash database:
insert into duplicate_hashs(hash,value) values($hash,$value)
while $no_more_duplicates != true
{
update into the duplicate hashs all the ones (1,2,3,4) where duplicates were found
}

Alec

Wow. A new guy.

I mean, holy shit.