Search This Blog

Thursday, May 5, 2016

Oracle on linux configurtion

we find out huge differences when retrieving huge amount of data from oracle , when it was installed on linux and on windows ( the windows was allot faster)

DBA:
 check the next articles and compare/add/fix the values that suggested by oracle for Linux  - you might find there what we are looking for https://docs.oracle.com/database/121/LTDQI/toc.htm#BHCCADGD
Have a look at this article http://www.dba-oracle.com/art_builder_tns.htm - there are some critical parameters to check (tcp.nodelay = yes ?????).

IT:
I suggest to improve the kernel for all our application server RedHat + zLinux with the necessary changes for best throughput , concentrate with the Network tuning
There are allot of resources how to improve it if you didn’t    (specially net.core.rmem_max , net.core.wmem_max)



enjoy 
Yaniv 

Wednesday, April 6, 2016

sqlcmd - command line to run sql scripts (batch mode)

sometimes the best option to get fast result is to use an exisitng queries tools, many option include formatting for csv files.



Use sqlcmd with Scripting Variables:

sqlcmd Utility