Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [eclipselink-users] query optimization for secondary keys 
Hi David,
Perhaps you can give an example of the kind of query that might give you an issue.
The way EclipseLink is going to handle queries for multiple entities is that unless specifically instructed not to go to the database, it will query the database. The query from the database will bring bring back information about all the appropriate entities including the fields that have been identified as identity fields in the entity. Those identity fields will be used for the cache lookup. 

-Tom

David Bernstein wrote:

Is there a generic way to have an EJB3 query take advantage of database
secondary composite keys?  If there's no generic way, then I'd accept an
EclipseLink-specific method, but it seems there ought to be a generic JPA
way to do this.
For example, suppose there were an SQL table create table su ( 
  su_id int primary key,
  ri    int not null,
  rt    int not null,
  unique ( ri, rt )
  );
where I want su_id to be the primary key, but I also want efficient access 
via the composite key (ri,rt) using EJB3/EclipseLink.
I think I may have a performance problem because the database is configured 
for fast access via the unique composite secondary key, but the cache would
be searched linearly (and the entire table may be sucked in?) when querying
via the secondary key.
The @Table annotation has a uniqueConstraints element, but it appears to me 
that it’s only used for schema generation, not query optimization.
It seems really basic to me to be able to do this so I think I must be 
missing something really obvious. My googling and manual reading haven't
helped much.  At the moment the only thing I can think of is to do a native
named query, but that would seem to bypass the persistence provider's cache. 
*Apologies in advance if anybody receives this message twice.*

Back to the top