A resource's initial lookup of its stack root id is inefficient

Bug #1677901 reported by Crag Wolfe
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
OpenStack Heat
In Progress
Medium
Zane Bitter

Bug Description

Resource objects already have a root_stack_id, so once they are set they can be looked up in constant time. The issue is that the first call to determine a resource's root stack id (when the resource is created) invokes https://git.openstack.org/cgit/openstack/heat/tree/heat/db/sqlalchemy/api.py?id=6abdb1#n751 , or D database reads where D is the depth of the resource. Or, for an entire tree we are doing ~ N log(N) extra DB reads to determine all the resources' root stack ID's.

Changed in heat:
assignee: nobody → Crag Wolfe (cwolfe)
status: New → In Progress
Rico Lin (rico-lin)
Changed in heat:
milestone: none → pike-3
Rico Lin (rico-lin)
Changed in heat:
milestone: pike-3 → pike-rc1
importance: Undecided → Low
Revision history for this message
Zane Bitter (zaneb) wrote :

Not sure why the bot missed it, but the patch is https://review.openstack.org/#/c/354621/

Changed in heat:
assignee: Crag Wolfe (cwolfe) → nobody
milestone: pike-rc1 → next
importance: Low → Medium
Changed in heat:
assignee: nobody → Zane Bitter (zaneb)
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.