Hello All,
Its Aurora postgresql database in AWS. We have a couple of custom tables created with some ~100 odd columns and required indexes and constraints created on them and that represents a certain business use case. Currently we don't have any data in them. Requirement is to test read and write queries performance for this sample use cases/tables. We need to evaluate both row by row and batch read/write performance. Also performance with concurrent read/write from multiple threads/sessions.
As i understand below things to be done here,
1)For write performance , the rows needs to be inserted from multiple sessions at same time, with required random values as per the data types i.e. Character, Number, date columns. And this needs to be tested for row by row insert and batched insert.
2)For the read performance test , the table first has to be populated with those ~100million rows. Then querying will happen on that table row by row and batched way from the concurrent session.
I am new to postgresql but mostly worked with Oracle, so I wanted to understand if the above can be achieved by creating a simple procedure or will a major effort be needed? And I saw a few blogs , pgbench to be used for concurrency tests. I want to know if this will still work on Aurora postgresql from intellij client worksheet.
Its Aurora postgresql database in AWS. We have a couple of custom tables created with some ~100 odd columns and required indexes and constraints created on them and that represents a certain business use case. Currently we don't have any data in them. Requirement is to test read and write queries performance for this sample use cases/tables. We need to evaluate both row by row and batch read/write performance. Also performance with concurrent read/write from multiple threads/sessions.
As i understand below things to be done here,
1)For write performance , the rows needs to be inserted from multiple sessions at same time, with required random values as per the data types i.e. Character, Number, date columns. And this needs to be tested for row by row insert and batched insert.
2)For the read performance test , the table first has to be populated with those ~100million rows. Then querying will happen on that table row by row and batched way from the concurrent session.
I am new to postgresql but mostly worked with Oracle, so I wanted to understand if the above can be achieved by creating a simple procedure or will a major effort be needed? And I saw a few blogs , pgbench to be used for concurrency tests. I want to know if this will still work on Aurora postgresql from intellij client worksheet.
Regards
Veem
