Dear ALL,

I am really thankful to  Bill King, Micha Bieber , Derrell for your valuable 
suggestions.

I was really confused that which way should I follow now, because I was sure 
that SQLite will work much better. when I got the suggestion about Firebird 
then again I went to the comparison page of SQLite and found that Firebird 
was also slow in many cases compared to SQLite. 

>From the DerrellÂ’s reply again I got hope that yes I can achieve that using 
SQLite.

Now I want to add some more information about my work may be you will add 
some suggestions.     

//--------------------------------------------------------------------
TEST 1  ( Disk DB )

Table Names     itm , HVH       
Number of Records :  itm   100,000 and HVH   less than 10,000

QUERY:
Select count(*) from itm, HVH where  itm .IDC = HVH.IDC
Elapse Time  : 54.359  (Sec)

//--------------------------------------------------------------------

TEST 2 ( Disk DB )

Table Names     itm , HVH       
Number of Records :  itm   5 Million  and HVH   less than 10,000

QUERY:
create index index1 on itm(IDC) 
Elapse Time  : 0.356  (Sec)


Select count(*) from itm,HVH where  itm.IDC=HVH.IDC
Elapse Time  : 64   (Sec)

//--------------------------------------------------------------------

TEST 3 ( Disk DB )

Table Names     itm     
Number of Records :  itm   20 Million  

QUERY:
Select count(*) from itm 
Elapse Time  : 133  (Sec)


//--------------------------------------------------------------------
 
Now Please have look at these results. What should I do now to improve the 
performance.
 Thanks and regards 
Manzoor Ilahi Tamimy



On Mon, 19 Jun 2006 09:47:03 -0400, Derrell.Lipman wrote
> Bill King <[EMAIL PROTECTED]> writes:
> 
> > Manzoor Ilahi Tamimy wrote:
> >
> >> We are Using SQLite for one of our project.
> >>
> >>The Database Size is  more than 500 MB.
> >>It contain one table and about 10 million Records.
> 
> > Err, for that size, I'd recommend going something heavier, like
> > firebird. This is not sqlite's solution domain in the slightest.
> 
> I'd have to differ on opinion here.  I have an sqlite database 
> that's at 6.9GB with about 40 million records, and it's working just 
> fine.  My query speed has not changed substantially as the database 
> has grown.
> 
> With sqlite, you *do* need to hand-optimize your queries.  Determine 
> which queries are slow, and consider splitting them into multiple 
> queries which generate temporary tables.  By doing this, you can 
> create indexes, as appropriate, on the temporary tables that make 
> the overall time to accomplish your goal much less than cramming it 
> all into a single query that is not so highly optimized.
> 
> Oh, and I'm using the old sqlite 2.8 series.  I expect I'd be 
> getting even better speed if I used the newer 3 series.
> 
> Cheers,
> 
> Derrell

Reply via email to