Re: Scalability fo Batch Operations in RAC

  • From: Mogens Nørrgaard <mln@xxxxxxxxxxxx>
  • To: oracle-l@xxxxxxxxxxxxx
  • Date: Tue, 27 Dec 2005 02:11:24 +0100

So the overhead associated with communication between two nodes is 50%? Sounds like bad scaling, but we've seen it worse :) We need the wait-profile, not the blocking and interconnect information.

Best regards,


Ravi_Kulkarni@xxxxxxxx wrote:

Are you using Parallelism (are you seeing parallel slaves being spawned on node 2 as seen in gv$px_session when batch job is running) ? DB /Table/index level ?
Try rerunning load by turning off Parallelism at each level.
Remember, default DOP (degree of parallelism) is a function of # instances, cpu_count and few parallel% parameters.

*From:* oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] *On Behalf Of *Sriram Kumar
*Sent:* Monday, December 26, 2005 9:50 AM
*Subject:* Scalability fo Batch Operations in RAC

Hi Gurus,
We are running a heavy batch operation of a banking application on RAC. Say the configuration is a 2 node 4 CPU Itanium2 box running <>.


Other related posts: