Hey hash

Discussion in 'LnS English Forum' started by Lavender, Sep 19, 2004.

Thread Status:
Not open for further replies.
  1. Lavender

    Lavender Registered Member

    Joined:
    Sep 19, 2004
    Posts:
    7
    I've been coming here for long time. And over this time I've learned a lot but never post any thread. Now its time to do so , I guess..
    Well, I struggled with the study of security..I reached the point where I need to know about Hash function..so any help will be appreciated..maybe someone can give me hints or websites to learn about it..any!!??

    by the way, is there is any other name for this kind of functions?

    nite nite
     
  2. gkweb

    gkweb Expert Firewall Tester

    Joined:
    Aug 29, 2003
    Posts:
    1,932
    Location:
    FRANCE, Rouen (76)
    Hi,

    I do not have links, but these hash functions have many names, they can be called message digest functions I think, or one way hash functions.

    The most known are MD5 and SHA family I think.
    The SHA family is composed of different functions, such as SHA-1, SHA256, SHA512 (there is others).

    You will have more answers on a more general forum at wilder.

    gkweb.
     
  3. Digger

    Digger Registered Member

    Joined:
    Oct 16, 2004
    Posts:
    7
    The following comes from DIAMONDCS "CryptoSuite" HELP File which may give you an idea of HASHing:

    What is a checksum

    A checksum is basically an extremely large number which represents a group of data. In its simplest form it is like referencing a larger set of data by a smaller set. If your name was Michael for instance, if someone called you Mike you would know they were referring to you even though they didn't use your full name. This is similar to how a checksum works.

    Checksums come in many different sizes, ranging from 1bit to 1024bit and anything over. Obviously the larger the checksum the more unique sets of data it can reference before "collisions" occur. What is a collision? A collision occurs when the same checksum or number is generated for 2 different sets of data.

    What the checksum does to generate a number given a set of data, is to go through ALL the data and while it is, do certain things. The simplest checksums simply add up each byte in the data. So if I had a 20byte piece of data where each byte contained the number 1, and I ran a simple adding checksum over it, it would give me the number 20. If I changed one of those bytes in the 20, to a 2, then the same checksum would now be 21. I would know there is a difference between the two sets of 20bytes simply by looking at the checksum. But what if in those 20 bytes, there was one 20, and nineteen 0's. It would also give me the checksum of 20, which would be a collision. Simple adding checksums are only used today when extreme speed is required, because they are very simple and produce a lot of collisions.

    Cryptographic checksums like SHA/HAVAL/MD5/TIGER/etc, use advanced mathematics principles to limit the amount of collisions and hence they are used today because of their high reliability.

    You can't work out what data a checksum actually refers to. This is because if a checksum is only 128bits(8 bytes) in length, any data over 128bits cannot be represented fully in 128bits. It's like trying to put more water into a cup that is already full. As in our previous checksum example, there is a lot of possibilities of the checksum of 20 appearing, hence we cannot work out which set of data actually gave us that checksum.
     
Thread Status:
Not open for further replies.