Hi Sergei,
another tip: instead of increasing the size of the data-file to >128K
you can also decrease the size of the buffer. I did something similar
for a test and (if I remember it correctly) it's done in sys/sysvars.cc
in the var. Sys_read_buff_size. I've set the entry for DEFAULT to 4K
and pl
Hi Sergei,
coming back to my observation.
With the text-file I already sent you in another thread here you can
also try to reproduce the effect I described here.
Here is how:
- modify the INSERT INTO TestBig-statement so that it will include much
more records into the table
- execute the statement
Hi Sergei,
thnkas for your reply.
In my tests I found that the first table-scan takes about 12 seconds,
the second table-scan takes about 25 seconds (and all folowing) scans
too).
When I call the function reinit_io_cache() all scans take about 12
seonds. But my implementation was quick and dirty...
Hi Robert,
#1: OS
I've only tested on Linux. But I was using the source-code of MariaDB
so I assume the effect will happen on Windows too.
#2: SQL-statement
this is the statement I've used for my test:
MariaDB [TestOpt]> select SQL_NO_CACHE B.PZN, B.ArtikelText from
TestBig B where exists ( sele
Hi, AugustQ!
On Jan 29, AugustQ wrote:
> Hi,
>
> by playing with the code I think I found something interesting.
>
> My environment: MariaDB 10.0.10, MyISAM-engine
>
> I played with a table-scan, no index is defined on this table. When I
> execute a SQL-statement that forces the server to do a
hum... did you checked only linux version, or windows too?
could you provide a SQL example?
2017-01-29 14:30 GMT-02:00 AugustQ :
> Hi,
>
> by playing with the code I think I found something interesting.
>
> My environment: MariaDB 10.0.10, MyISAM-engine
>
> I played with a table-scan, no index is
6 matches
Mail list logo
