On 07/10/2011 02:31 PM, Samuel Gendler wrote:
> What about partitioning tables by tenant id and then maintaining
> indexes on each partition independent of tenant id, since constraint
> exclusion should handle filtering by tenant id for you. That seems
> like a potentially more tolerable variant of #5 How many tenants are
> we talking about? I gather partitioning starts to become problematic
> when the number of partitions gets large.
I thought I had replied... Apparently I didn't.
The database can grow in two dimensions: The number of tenants and the
number of rows per tenant.
We have many tenants with relatively little data and a few with a lot of
data. So the number of tenants
is known ahead of time and might be 1000's.
In response to
pgsql-performance by date
|Next:||From: lars||Date: 2011-07-16 00:21:28|
|Subject: Re: UPDATEDs slowing SELECTs in a fully cached database|
|Previous:||From: Josh Berkus||Date: 2011-07-15 19:46:16|
|Subject: Re: Hardware advice for scalable warehouse db|