| << Back to previous view |
|
[QB-1899] Constraining the size of EC2 cloud
|
|
| Status: | Resolved |
| Project: | QuickBuild |
| Component/s: | None |
| Affects Version/s: | 5.1.6 |
| Fix Version/s: | 5.1.25 |
| Type: | Improvement | Priority: | Major |
| Reporter: | Lukasz Guminski | Assigned To: | Unassigned |
| Resolution: | Fixed | Votes: | 1 |
| Remaining Estimate: | Unknown | Time Spent: | Unknown |
| Original Estimate: | Unknown | ||
| Description |
|
To make a cloud profile usable for production, we need to be able to constrain the size of the cloud. Sometimes we have several hundreds requests in the queue, and letting QB decide on new resources, could become very costly.
You could apply Amazon scaling concepts, where an autoscaling group accepts the following parameters: - minimal capacity - maximal capacity - desired capacity - cooldown period http://docs.aws.amazon.com/AutoScaling/latest/DeveloperGuide/AS_Concepts.html |
| Comments |
| Comment by Robin Shen [ 06/May/14 11:33 AM ] |
| The already exist "idleTimeToTerminate" property can play the role of "cooldown" period. For "maximum capacity", we will enhance the cloud profile in next patch release to support such property. Builds requiring more capacity of the same profile will be put into queue waiting. |