which is more scalable for the database? - Mailing list pgsql-general

From Timasmith
Subject which is more scalable for the database?
Date
Msg-id 1173268012.906695.70300@64g2000cwx.googlegroups.com
Whole thread Raw
Responses Re: which is more scalable for the database?  (Richard Huxton <dev@archonet.com>)
Re: which is more scalable for the database?  (Shane Ambler <pgsql@Sheeky.Biz>)
List pgsql-general
Suppose I have a database table with 20 fields which are lookups to a
single table.

configtable(configtable_id, a_field, something_lookup_id,
another_lookup_id, ...)
lookup(lookup_id, value, description, ...)


what is going to be faster to map the rows to an Object which needs
the 'value' for every field ending in lookup_id


a) View


select c.configtable_id, l1.value as something_lookup, l2.value as
another_lookup
from configtable c,
       lookup l1,
       lookup l2
where c.something_lookup_id = l1.lookup_id
    and c.another_lookup_id = l2.lookup_id


foreach row
   map values to object
end


b) Cache all lookup values and populate


select c.* from configtable


foreach row
   map values to object
   if lookup_id
       find value from hashtable and map value to object
   endif
end


It seems that the latter *might* be better to scale outward better,
as
you could add application servers to do the caching/mapping and you
only select from a single table?


pgsql-general by date:

Previous
From: RPK
Date:
Subject: Determine users and roles
Next
From: "Reuven M. Lerner"
Date:
Subject: Re: Database slowness -- my design, hardware, or both?