Hello,
My friend and I were arguing about different way to make postgresql run faster. We have argued about this several times and we can not reach an agreeable solution.
I would appreciate it if you guys can weigh in with your thoughts.
Assuming that the query is a simple select query running on a large table with over a million records. The table would probably indexed several times, and tere is no calculation needed in this query.
If the machine that postgresql is running on is a SMP machine with around a couple of gigs of ram, and the query runs slowly. What would make the query run faster?
I believe that adding a faster hard drive such as scsi 160's on raid 0 or 1 would be a better solution. He believes that adding more ram would be a better solution. In my opinion if we have two hard drives acessing records in a database it will be fater than one.
What do you guys thing? Are we both wrong, what would make this query run faster?
Your answers based on practical exprience is welcomed. I appreciate answers based in theory, but as I have found out in my few years in the comptuer buisness, theory rarely translates well to the real world.
What do you guys thing? Are we both wrong, what would make this query run faster?
My friend and I were arguing about different way to make postgresql run faster. We have argued about this several times and we can not reach an agreeable solution.
I would appreciate it if you guys can weigh in with your thoughts.
Assuming that the query is a simple select query running on a large table with over a million records. The table would probably indexed several times, and tere is no calculation needed in this query.
If the machine that postgresql is running on is a SMP machine with around a couple of gigs of ram, and the query runs slowly. What would make the query run faster?
I believe that adding a faster hard drive such as scsi 160's on raid 0 or 1 would be a better solution. He believes that adding more ram would be a better solution. In my opinion if we have two hard drives acessing records in a database it will be fater than one.
What do you guys thing? Are we both wrong, what would make this query run faster?
Your answers based on practical exprience is welcomed. I appreciate answers based in theory, but as I have found out in my few years in the comptuer buisness, theory rarely translates well to the real world.
What do you guys thing? Are we both wrong, what would make this query run faster?