most of the db applications tend to heavily perform range queries which
btrees are excellent for.
range queries are something like
select * from my_table where age > 30 etc
just for fetching exact values (==) hash tables are still faster by using
hash functions like md5. amount of data doesnt matter
with caching and disk-based hashes.
even if u use a hash function which has collision, by using multiple levels
of hashes they would still beat btrees. this is my opinion.
everything depends on the operations u perform on the data structure.
anything other than == opeartion ,hash table is simply a waste and hence not
used in dbs. (and btree for == is reasonbly faster if not as much as
hashes).


On 4/18/07, C++4LifePuta <[EMAIL PROTECTED]> wrote:
>
>
> A programmer friend of mine told me that research was done and for
> database applications, B-Trees were faster on average as opposed to
> hashes.
>
> I thought about this for a minute and could only come up with maybe
> the amount of data they are storing is so large that a resonably
> unique hash function could not always be derived resulting in frequent
> collisions.
>
> Any thoughts on this?
> <first time poster>
>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Algorithm Geeks" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/algogeeks
-~----------~----~----~----~------~----~------~--~---

Reply via email to