[
https://issues.jboss.org/browse/TEIID-2461?page=com.atlassian.jira.plugin...
]
Steven Hawkins resolved TEIID-2461.
-----------------------------------
Resolution: Done
Switched the logic to simply slow-down, but not altogether stop processing when the
forward-only buffer limit is hit. The slow down is based upon source connections being
held (longer delay if nothing is held). This ensures a more timely completion, but will
likely see increased disk use.
There is currently no easy approach to determine which resource is more valuable (source
connection vs. disk) and whether the client will ever work down the buffer, so holding
source connections indefinitely is not a good option.
Other possibilities would be:
- a configurable max size of all tuplebuffers or just the output buffer (to prevent
runaway disk consumption)
- encourage the use of a local connection that is using the calling thread (which will
always process on demand and not build an output buffer for forward only)
- allow for a synch processing hint / execution mode to allow the user query to request
that the query is processed on demand.
Better determine when to block on the output buffer
---------------------------------------------------
Key: TEIID-2461
URL:
https://issues.jboss.org/browse/TEIID-2461
Project: Teiid
Issue Type: Quality Risk
Components: Query Engine
Reporter: Steven Hawkins
Assignee: Steven Hawkins
Fix For: 8.4
The output buffer size limitation helps prevent forward only results from unnecessarily
consuming resources, but it also may slow down the processing of otherwise already
buffered results - which will hold those and other downstream resources unnecessarily. We
should examine the plan to see if this is possible before allowing the output buffer to be
size limited.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:
http://www.atlassian.com/software/jira