After 4 days and a half, my program finished last night. The 1045175762 rows costed the Java program 109 hours to load. The speed is about 9.6 million per hour.

I didn't use multiple insert statements and "LOAD DATA INFILE", but only "insert into" statement. The Java program reads one line from the source file, and then execute an "insert" statement.

The configuration of my server is as follows:
Two Intel Xeon 2.8GHz CPUs
3GB memory
Two RAID 5 of SATA disks

The huge file locates on a RAID, the MySQL server datadir locates on another 
RAID. The two RAIDs link to the server by Ultra SCSI 320 interface.





Gary Richardson wrote:

Are you doing single insert statements, multiple insert statements or
LOAD DATA INFILE statements?

On 12/4/05, Chenzhou Cui <[EMAIL PROTECTED]> wrote:
Dear  MySQL fans,

I have a 260 GB huge file with 1045175762 rows. Two weeks ago, I wrote a
Java program to read the huge plain text file into MySQL 4.1.12.  300
million of rows could be loaded in one day.

Last Thursday, I updated the MySQL to 5.0.16 and then run the same
program. Only 225 million of rows can be loaded in one day.

In additional to the version difference, the MySQL 4.1.12 was compiled
from SRPM packages from RedHat Updates, while the MySQL 5.0.16 is
installed directly using the MySQL binary RPM packages for Redhat AS 4.

Any information and suggestion are welcome.

Regards,
Chenzhou CUI


--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]




--
============================================================
Chenzhou Cui (China-VO Project) National Astronomical Observatory | Tel: (8610)64841695
Chinese Academy of Sciences       | FAX: (8610)64878240
Datun Road 20A, Chaoyang District | Email: [EMAIL PROTECTED]
Beijing 100012, China             | WWW: www.lamost.org/~cb
============================================================



--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]

Reply via email to