----- Original Message ---- > From: [EMAIL PROTECTED] > <[EMAIL PROTECTED]> wrote: > > > > One question though: are the file access "sorted", so that seeks are > > minimised when performing a transaction (making the assumption that > > the file is not fragmented on disk)? > >
> I'm not sure what you are asking. Can you restate your question? Basically, the db file is accessed with seek + read/write operations. Given a set of such operations, it can be very beneficial to reorder those operations so that the file is accessed going from the begining to the end of the file (and not pure random). This is a problem mostly for reads (writes are reordered when doing a combination of async writes + flush at the end). I am guessing that given the nature of sql (a lot of iterations through keys/indexes), it might be possible to create batches of disk i/o queries: create a table for the next N reads (with offset) and do those reads sorted by offset. In the graphics world (where I come from), it's common practice to at least do prefetching of data (send in batch on the bus), and I believe that the benefits would be similar here. Nicolas
