[
https://issues.apache.org/jira/browse/OPENJPA-1991?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13786149#comment-13786149
]
Vermeulen commented on OPENJPA-1991:
------------------------------------
This definitely has my vote. When I use setMaxResults the generated SQL doesn't
even use TOP and fetches all results and filters in memory. A not very optional
workaround I can use is selecting id's first using setMaxResults and then
select the entities with these id's. This means all id's are still fetched, but
in my use case performance is acceptable.
> Support ROW_NUMBER pagination on MS SQL Server 2005
> ---------------------------------------------------
>
> Key: OPENJPA-1991
> URL: https://issues.apache.org/jira/browse/OPENJPA-1991
> Project: OpenJPA
> Issue Type: Improvement
> Components: jdbc
> Affects Versions: 2.1.0
> Reporter: Maia Kozheva
> Attachments: SQLServer2005Dictionary.java
>
>
> The current implementation of SQLServerDictionary in OpenJPA only supports
> pagination based on TOP (offset+limit), which fetches more rows than
> necessary and thus degrades performance in comparison to DBMSes that support
> both limit and offset specification, like MySQL.
> SQL Server 2005 supports a ROW_NUMBER function that can be used to restrict
> the result list to the necessary slice, taking both limit and offset into
> account. I would like OpenJPA to support this syntax.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
