Thread: DB Analysis
Hi,
I need to analyse a development DB which runs a whole slew of queries.
It is Postgres 8.2 running on Fedora Core 8 2.6.23.1-42 on a 64 bit AMD Athlon(tm) 64 X2 Dual Core Processor 5000+ machine with 4GB DDR2 800Mhz RAM.
I need to figure out how long queries take, should take, etc etc.
Can someone point me in the right direction - either a resource that I can read and get an idea from or a tool I can run.
For example, if someone is running a piece of code that affects say, 1000 records in a table, how long should it take ? Is it possible to answer that question ? If so, how would I analyse it ?
I need to analyse a development DB which runs a whole slew of queries.
It is Postgres 8.2 running on Fedora Core 8 2.6.23.1-42 on a 64 bit AMD Athlon(tm) 64 X2 Dual Core Processor 5000+ machine with 4GB DDR2 800Mhz RAM.
I need to figure out how long queries take, should take, etc etc.
Can someone point me in the right direction - either a resource that I can read and get an idea from or a tool I can run.
For example, if someone is running a piece of code that affects say, 1000 records in a table, how long should it take ? Is it possible to answer that question ? If so, how would I analyse it ?
-- Regards, Arjun Datta
On Thu, 2008-08-07 at 14:40 -0400, Arjun Datta wrote: > > For example, if someone is running a piece of code that affects say, > 1000 records in a table, how long should it take ? Is it possible to > answer that question ? If so, how would I analyse it ? As for what each query should do.. I'm not sure that this question has an answer. It would relate entirely to your database schema and how you write your queries, and I am not aware of any benchmarks or metrics for this kind of information. On a side note, there's plenty of resources as to what each tuning parameter does and what the optimals are for many situations. -Mark