Database speed comparison

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Database speed comparison

Cecil Westerhof-5
I ran across this page:
    https://sqlite.org/speed.html

It is a 'bit' outdated. (It uses 2.7.6.) Where can I find the scripts? And
would it be a lot of work to update them and regularly rerun them? If it is
manageable I would not mind to provide regularly updates.

--
Cecil Westerhof
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
Reply | Threaded
Open this post in threaded view
|

Re: Database speed comparison

Andy-3
I like to compare these tests on my computer. My first attempt to Sqlite
works but a lot of slower and I don't know why, maybe due to inserting
blobs? But blobs are small, about 10 kB.

pt., 10 sty 2020 o 19:37 Cecil Westerhof <[hidden email]>
napisał(a):

> I ran across this page:
>     https://sqlite.org/speed.html
>
> It is a 'bit' outdated. (It uses 2.7.6.) Where can I find the scripts? And
> would it be a lot of work to update them and regularly rerun them? If it is
> manageable I would not mind to provide regularly updates.
>
> --
> Cecil Westerhof
> _______________________________________________
> sqlite-users mailing list
> [hidden email]
> http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
>
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
Reply | Threaded
Open this post in threaded view
|

Re: Database speed comparison

Simon Slavin-3
In reply to this post by Cecil Westerhof-5
Are you aware of

<https://www.sqlite.org/src/file/test/speedtest1.c>

?  Now all you have to do is write comparative tests for other SQL implementations.  However, from what I've seen, realistic speed tests tend to come down to cache sizes.
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
Reply | Threaded
Open this post in threaded view
|

Re: Database speed comparison

Edward Lau
In reply to this post by Cecil Westerhof-5
I would be nice to have that page updated and contrast the old and new performance of SQLite.


-----Original Message-----
From: Cecil Westerhof <[hidden email]>
To: SQLite mailing list <[hidden email]>
Sent: Fri, Jan 10, 2020 10:37 am
Subject: [sqlite] Database speed comparison

I ran across this page:
    https://sqlite.org/speed.html

It is a 'bit' outdated. (It uses 2.7.6.) Where can I find the scripts? And
would it be a lot of work to update them and regularly rerun them? If it is
manageable I would not mind to provide regularly updates.

--
Cecil Westerhof
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
Reply | Threaded
Open this post in threaded view
|

Re: Database speed comparison

Andy-3
In reply to this post by Simon Slavin-3
These tests are a lot faster than calling Sqlite in my program, I must use
the same convention: for example first prepare, next sqlite3_bind_int64,
sqlite3_bind_text and sqlite3_step. My calling was very not optimal.

pt., 10 sty 2020 o 20:13 Simon Slavin <[hidden email]> napisał(a):

> Are you aware of
>
> <https://www.sqlite.org/src/file/test/speedtest1.c>
>
> ?  Now all you have to do is write comparative tests for other SQL
> implementations.  However, from what I've seen, realistic speed tests tend
> to come down to cache sizes.
> _______________________________________________
> sqlite-users mailing list
> [hidden email]
> http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
>
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
Reply | Threaded
Open this post in threaded view
|

Re: Database speed comparison

Andy-3
I had sqlite3_prepare_v2 without sqlite3_finalize and sqlite3_close not
closes multiple databases.

sob., 11 sty 2020 o 07:17 Andy <[hidden email]> napisał(a):

> These tests are a lot faster than calling Sqlite in my program, I must use
> the same convention: for example first prepare, next sqlite3_bind_int64,
> sqlite3_bind_text and sqlite3_step. My calling was very not optimal.
>
> pt., 10 sty 2020 o 20:13 Simon Slavin <[hidden email]> napisał(a):
>
>> Are you aware of
>>
>> <https://www.sqlite.org/src/file/test/speedtest1.c>
>>
>> ?  Now all you have to do is write comparative tests for other SQL
>> implementations.  However, from what I've seen, realistic speed tests tend
>> to come down to cache sizes.
>> _______________________________________________
>> sqlite-users mailing list
>> [hidden email]
>> http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
>>
>
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
Reply | Threaded
Open this post in threaded view
|

Re: Database speed comparison

Andy-3
And huge speedup by exec("BEGIN") myloop exec("COMMIT")

sob., 11 sty 2020 o 07:30 Andy <[hidden email]> napisał(a):

> I had sqlite3_prepare_v2 without sqlite3_finalize and sqlite3_close not
> closes multiple databases.
>
> sob., 11 sty 2020 o 07:17 Andy <[hidden email]> napisał(a):
>
>> These tests are a lot faster than calling Sqlite in my program, I must
>> use the same convention: for example first prepare, next
>> sqlite3_bind_int64, sqlite3_bind_text and sqlite3_step. My calling was very
>> not optimal.
>>
>> pt., 10 sty 2020 o 20:13 Simon Slavin <[hidden email]> napisał(a):
>>
>>> Are you aware of
>>>
>>> <https://www.sqlite.org/src/file/test/speedtest1.c>
>>>
>>> ?  Now all you have to do is write comparative tests for other SQL
>>> implementations.  However, from what I've seen, realistic speed tests tend
>>> to come down to cache sizes.
>>> _______________________________________________
>>> sqlite-users mailing list
>>> [hidden email]
>>> http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
>>>
>>
_______________________________________________
sqlite-users mailing list
[hidden email]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users