Re: select count() out of memory

> tfinneid(at)student(dot)matnat(dot)uio(dot)no writes:
>>> In that case, why use partitions at all? They are simple independent
>>> tables.
>
>> For two reasons,
>> - the data logically belongs together.
>> - because its more practical to create tables as childs of a parent
>> table
>> than as independent tables.
>> - changes to the table is applied to all partitions, and prohibits
>> tables with different dd.
>> - performing the create operation does not require the source code to
>> contain the ddl of the parent table.
>
> In other words, you really should have only one table; they aren't
> independent. What you need to do is dial down your ideas of how many
> partitions are reasonable to have.

Yes, but no. Each partition represents a chunk of information on a
discrete timeline. So there is no point in grouping it all into a single
table, because the access pattern is to only access data from a specific
point in time, i.e. a single partition, usually the latest. Since the
amount of data is so big, approx 3MB per second, and each partition needs
to be indexed before the clients start reading the data (in the same
second). I find its better to use partitions, even though I am not
actually using it.

regards

thomas