Re: 500 times slower
От | Dave Page |
---|---|
Тема | Re: 500 times slower |
Дата | |
Msg-id | E7F85A1B5FF8D44C8A1AF6885BC9A0E4528A0B@ratbert.vale-housing.co.uk обсуждение исходный текст |
Ответ на | 500 times slower (Karol Szkudlarek <karol@mikronika.com.pl>) |
Список | pgsql-odbc |
> -----Original Message----- > From: Karol Szkudlarek [mailto:karol@mikronika.com.pl] > Sent: 09 February 2005 14:06 > To: Dave Page > Cc: pgsql-odbc@postgresql.org > Subject: Re: [ODBC] 500 times slower > > Hi Dave! > > In my test case I read LOB of 10 megabytes size. > So then in the function (with 4kB SOCK_BUFFER_SIZE): > SOCK_get_next_byte(SocketClass *self) > below line (socket.c): > > self->buffer_filled_in = recv(self->socket, (char *) self->buffer_in, > self->buffer_size, 0); > > runs in the following schema: 0msec, 200msec, 0msec, 200msec... etc. > So above times suggest delayed ACK timer (200msec). OK, that figures. I (and others I've asked) cannot see a reason not to increase the buffers size other than the extra memory that'll be used (which is trivial). Anyone disagree before I do so? Regards, Dave.
В списке pgsql-odbc по дате отправления: