Am 10.02.2010 23:17, schrieb Simon Slavin:
> But that's true only if you're running a SELECT which actually uses
> that column and only that column to do the searching. Which is why I
> asked that question earlier on in this thread.
> Simon.
>
The implementation of sqlite uses a B+Tree for
On Wed, Feb 10, 2010 at 10:17:28PM +, Simon Slavin scratched on the wall:
>
> On 10 Feb 2010, at 7:51pm, Ibrahim A wrote:
>
> > An "INTEGER PRIMARY KEY" is at least twice as fast as another type of
> > PRIMARY KEY,
> > the reason is based on the implementation of the engine. An integer
> >
On 10 Feb 2010, at 7:51pm, Ibrahim A wrote:
> An "INTEGER PRIMARY KEY" is at least twice as fast as another type of
> PRIMARY KEY,
> the reason is based on the implementation of the engine. An integer
> primary key substitutes the rowid column of a table.
But that's true only if you're
Am 10.02.2010 18:19, schrieb Alberto Simões:
> Supose a table with a key that is a string (say, words from 1 to 10
> characters) or a table with a key of integers.
>
> How different is the efficiency on fetching one record on these tables?
>
>
If you look into the documentation for "create
Sure (com certeza!), because it depends on the hardware and software of your
target platform.
2010/2/10 Alberto Simões
> On Wed, Feb 10, 2010 at 5:42 PM, Virgilio Fornazin
> wrote:
> > I think you should be asking 'How fast is SQLite locating a
On Wed, Feb 10, 2010 at 5:42 PM, Virgilio Fornazin
wrote:
> I think you should be asking 'How fast is SQLite locating a key in a integer
> column index vs a string index'...
>
> Generally, integer keys are faster in key lookups than string keys, because
> comparing a
I think you should be asking 'How fast is SQLite locating a key in a integer
column index vs a string index'...
Generally, integer keys are faster in key lookups than string keys, because
comparing a integer value is a
single CMP CPU instruction versus a more-complicated string comparison (that
On 10 Feb 2010, at 5:19pm, Alberto Simões wrote:
> I know I can benchmark myself this question, but I am sure somebody
> did that already.
>
> Supose a table with a key that is a string (say, words from 1 to 10
> characters) or a table with a key of integers.
>
> How different is the
Howdy SQLite users,
I know I can benchmark myself this question, but I am sure somebody
did that already.
Supose a table with a key that is a string (say, words from 1 to 10
characters) or a table with a key of integers.
How different is the efficiency on fetching one record on these tables?
9 matches
Mail list logo