Re: [sqlite] Performance problems potentially related to index scalability

2007-02-09 Thread Joe Wilson
> I understand that I'm experiencing thrashing, but I'm curious as to why I'm > experiencing it. My understanding is that the row_id PRIMARY KEY column has > an index on it by default. Inserting into this index doesn't cause > thrashing, but inserting into the secondary index does. I hoped that >

Re: [sqlite] Performance problems potentially related to index scalability

2007-02-09 Thread Stan Bielski
On 2/8/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: "Stan Bielski" <[EMAIL PROTECTED]> wrote: > > Anyone care to take a guess what's going on here? The problem is that you are thrashing. The working set on the database file is exceeding the amount of memory that your OS has set aside for

Re: [sqlite] Performance problems potentially related to index scalability

2007-02-08 Thread drh
"Stan Bielski" <[EMAIL PROTECTED]> wrote: > > Anyone care to take a guess what's going on here? The problem is that you are thrashing. The working set on the database file is exceeding the amount of memory that your OS has set aside for disk cache. The best solution, if possible, is to insert

Re: [sqlite] Performance problems potentially related to index scalability

2007-02-08 Thread Stan Bielski
On 2/8/07, Griggs, Donald <[EMAIL PROTECTED]> wrote: Regarding: "...Next, I add in an index on [one] of my numeric columns. ... This is where the problem comes in." You may already know this, but if you can tolerate waiting until the loads are complete before adding the index, the total

RE: [sqlite] Performance problems potentially related to index scalability

2007-02-08 Thread Griggs, Donald
Regarding: "...Next, I add in an index on [one] of my numeric columns. ... This is where the problem comes in." You may already know this, but if you can tolerate waiting until the loads are complete before adding the index, the total performance should be better.