RE: [HACKERS] [hackers]development suggestion needed

Поиск
Список
Период
Сортировка
От Ansley, Michael
Тема RE: [HACKERS] [hackers]development suggestion needed
Дата
Msg-id 1BF7C7482189D211B03F00805F8527F748C446@S-NATH-EXCH2
обсуждение исходный текст
Ответы Re: [HACKERS] [hackers]development suggestion needed  (Brian E Gallew <geek+@cmu.edu>)
Список pgsql-hackers
>> > So, what's the deal, here...is the goal the Big Time or not?
>> 
>> If it means adopting one vendors concept of what the world should look
>> like...not.
>> 
>> I *hate* the way Oracle sets up tablespaces ... where i have to
>> pre-guess
>> the size of my data and allocate space accordingly...what if my table
>> never does reach that critical mass?  I've just wasted X meg of space
>> ...
I go with you on the size thing, but I still think it's not a bad idea to be
able to determine WHERE your data goes, at least down to a table/index/etc.
level.

>> I hate the way that Oracle starts up something like 4 processes for
>> every
>> database, when the server is started...
Well, Oracle doesn't see a database quite the way we do.  At least, not the
way we use it.  We tend to have multiple schemas in a single 'instance', or
database.  These schemas are actually defined per user.  This is possible on
Postgres, but people just don't do it.  So although four processes are
started for each instance, we only have two instances running on our main
dev server, even though there are about twenty-five schemas on it.

<snip>

>> If that happens to follow what one vendor happens to have done as far >>
as
>> their implementation, great...but there has been no conscious effort to
>> do
>> so that I'm aware of...
Cool.  PostgreSQL is a vendor.

>> Just look at the whole OUTER JOIN issue ... *shrug*
>> 
>> I *like* the fact that we come up with original ideas/solutions to
>> problems, that make use of *existing* technologies ...
And move to new technologies a lot quicker than any other product.

>> I liked the thread about moving indexes and tables to seperate file
>> systems, and hope we can implement something that will make it 
>> something
>> that does't require 'ln's, but I definitely don't like Oracle's way of
>> doing things ...
Yes, that's about the sum of it.  Why not the links?  I think that it's an
elegant way of designing the whole thing.  Only the system table that stores
the 'tablespace' directories will even have a hard path in it.  For the
rest, everything works in the main database directory (which could be
considered the SYSTEM tablespace).

MikeA


В списке pgsql-hackers по дате отправления:

Предыдущее
От: "Ansley, Michael"
Дата:
Сообщение: RE: [HACKERS] Re: Informix and OUTER join syntax
Следующее
От: Alfred Perlstein
Дата:
Сообщение: Re: [HACKERS] Revised nonblocking patches + quasi docs