Simultateous connection from AP server (Pega Platform 8.7) to Oracle DB
Hi, Pega Support I'm planning to deploy Pega platform by installing it onto RHEL and planning to deploy oracle 19c for the system. There's information that I require for planning. -The amount of simultaneous connections from Application Server(Pega App) to Database(oracle) =>3,700 users Access the Pega App simultaneously. And I'd like to know how many simultaneous connections will occur from App Server to DB in this situation. I'd appreciate if you could answer as soon as possible. Regards, Yukiko
Accepted Solution
Updated: 7 Jul 2022 23:13 EDT
JPMorgan Chase
IN
@YukikoM6 That you have to determine by doing application load testing.
It cannot be a straight forward number as such. based on your application business logic, number of database round trips, operations etc. lot of things come into picture.
During load testing identify the peak at which database connections start's entering into wait mode and resources starts degrading, that is the threshold. Taking that a base line identify the number of users at which this threshold is met, now replicate this infrastructure configuration proportionately to accomodate the users as per your application need
JPMorgan Chase
IN
While you establish the database connection through data source of your application server. The maximum number of connections that are made to database is controlled by connectionPoolSize settings in the data source configuration.
For e.g. if it is configured 10 then only 10 pool size all the transactions happen.
Rakuten Card Co., Ltd.
JP
@SriHarsha Anika Thank you for your reply and kind support
I understand that max connection numbers to database depends on connection PoolSize of database. I would like to set the poolsize so that all the connections to be accepted. In this case(3,700 users Access the Pega App simultaneously), what poolsize shold I set to data source configuration?
Yukiko Mitani
Accepted Solution
Updated: 7 Jul 2022 23:13 EDT
JPMorgan Chase
IN
@YukikoM6 That you have to determine by doing application load testing.
It cannot be a straight forward number as such. based on your application business logic, number of database round trips, operations etc. lot of things come into picture.
During load testing identify the peak at which database connections start's entering into wait mode and resources starts degrading, that is the threshold. Taking that a base line identify the number of users at which this threshold is met, now replicate this infrastructure configuration proportionately to accomodate the users as per your application need