NULL
>13VerifyCookie 1 2 000 NULL
>14TableLock 1 2 0 t 00 NULL
>15Goto 0 3 000 NULL
>
>-Urspr?ngliche Nachricht-----
>Von: sqlite-users-bounces at mailinglists.sq
14. Dezember 2015 13:54
>An: sqlite-users at mailinglists.sqlite.org
>Betreff: [sqlite] Is rowid the fastest?
>
>hi,
> create a table using the following sql.
> CREATE TABLE t(x INTEGER PRIMARY KEY ASC, y, z);
> So the x is the alias of the rowid. Retrieving records by rowid aroun
hi,
create a table using the following sql.
CREATE TABLE t(x INTEGER PRIMARY KEY ASC, y, z);
So the x is the alias of the rowid. Retrieving records by rowid around twice
as fast as other indexs values.
Because of the x is the alias of rowid, so retrieving records by x is also as
??? wrote:
> CREATE TABLE t(x INTEGER PRIMARY KEY ASC, y, z);
>
> In order to testing the efficiency. After I create index on t(x).
> And this give me a suprise, now I retrieve by x , I found that its is
> faster than its before.
An index entry is smaller than a table row, so if you
An: SQLite mailing list
Betreff: [sqlite] Is rowid the fastest?
You said that "You are probably falling into the cache effect trap again. There
is no point in indexing on the primary key, it only wastes space and CPU cycles
".
I do not agree with you. let me tell you why.
Before I ret
Auftrag von ???
Gesendet: Montag, 14. Dezember 2015 13:54
An: sqlite-users at mailinglists.sqlite.org
Betreff: [sqlite] Is rowid the fastest?
hi,
create a table using the following sql.
CREATE TABLE t(x INTEGER PRIMARY KEY ASC, y, z);
So the x is the alias of the rowid. Retrieving records
6 matches
Mail list logo