CPU to Memory weight based algorithm to order cluster #10997
CPU to Memory weight based algorithm to order cluster #10997yadvr merged 11 commits intoapache:mainfrom
Conversation
host.capacityType.to.order.clusters config will support new algorithm: COMBINED which will work with host.capacityType.to.order.clusters.cputomemoryweight and capacity will be computed based on CPU and memory both and using weight factor
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #10997 +/- ##
============================================
+ Coverage 16.56% 16.58% +0.01%
- Complexity 14010 14036 +26
============================================
Files 5758 5758
Lines 511578 511717 +139
Branches 62192 62216 +24
============================================
+ Hits 84756 84870 +114
- Misses 417350 417374 +24
- Partials 9472 9473 +1
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch. |
|
@blueorangutan package |
|
@sudo87 a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✖️ el8 ✖️ el9 ✖️ debian ✖️ suse15. SL-JID 13745 |
|
@blueorangutan package |
|
@sudo87 a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 13751 |
|
@sudo87 will you create a doc PR for this as well? |
Yes @DaanHoogland, doc pr will be needed for this change. |
engine/schema/src/main/java/com/cloud/capacity/dao/CapacityDaoImpl.java
Outdated
Show resolved
Hide resolved
server/src/main/java/com/cloud/agent/manager/allocator/impl/FirstFitAllocator.java
Outdated
Show resolved
Hide resolved
server/src/main/java/com/cloud/agent/manager/allocator/impl/FirstFitAllocator.java
Outdated
Show resolved
Hide resolved
|
@blueorangutan package |
|
@sudo87 a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 13821 |
|
@blueorangutan test |
|
@sudo87 a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests |
|
[SF] Trillian test result (tid-13554)
|
engine/components-api/src/main/java/com/cloud/configuration/ConfigurationManager.java
Outdated
Show resolved
Hide resolved
|
@blueorangutan package |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 14022 |
There was a problem hiding this comment.
Pull Request Overview
This PR adds support for a new "COMBINED" ordering mode that ranks clusters, pods, and hosts by a weighted sum of CPU and memory usage, introduces a config key for the CPU-to-memory weight, updates the DAO and schema to fetch both capacity types, and adds related unit tests.
- Added
COMBINEDoption andHostCapacityTypeCpuMemoryWeightconfig key - Extended
FirstFitPlanner,FirstFitAllocator, andCapacityDaoto compute and sort by combined metrics - Updated tests, configuration enum, and DB migration for the new behavior
Reviewed Changes
Copilot reviewed 13 out of 13 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
| server/src/main/java/com/cloud/deploy/FirstFitPlanner.java | Added helpers for combined CPU/memory ordering and refactored cluster/pod listing |
| server/src/main/java/com/cloud/agent/manager/allocator/impl/FirstFitAllocator.java | Added combined-capacity ordering and helper methods for hosts |
| engine/components-api/src/main/java/com/cloud/configuration/ConfigurationManager.java | Introduced HostCapacityTypeCpuMemoryWeight config key |
| server/src/main/java/com/cloud/configuration/Config.java | Updated Config.HostCapacityTypeToOrderClusters to include COMBINED |
| engine/schema/src/main/resources/META-INF/db/schema-42010to42100.sql | Migrated config description to mention COMBINED |
| engine/schema/src/main/java/com/cloud/capacity/dao/CapacityDao.java | Updated DAO interface signatures for capacity listing |
| engine/schema/src/main/java/com/cloud/capacity/dao/CapacityDaoImpl.java | Added methods to list capacities by types and removed obsolete signatures |
| server/src/test/java/com/cloud/vm/FirstFitPlannerTest.java | Added tests for combined ordering on clusters and pods |
| server/src/test/java/com/cloud/agent/manager/allocator/impl/FirstFitAllocatorTest.java | Added tests for combined ordering on hosts |
| engine/schema/src/test/java/com/cloud/capacity/dao/CapacityDaoImplTest.java | Extended DAO tests for new listing methods |
Comments suppressed due to low confidence (4)
server/src/main/java/com/cloud/deploy/FirstFitPlanner.java:526
- Missing import for ArrayList: add 'import java.util.ArrayList;' to ensure this compiles.
return new Pair<>(new ArrayList<>(podsByCombinedCapacities.keySet()), podsByCombinedCapacities);
server/src/main/java/com/cloud/deploy/FirstFitPlanner.java:564
- Missing import for ArrayList: add 'import java.util.ArrayList;' to the top of the file.
return new Pair<>(new ArrayList<>(clusterByCombinedCapacities.keySet()), clusterByCombinedCapacities);
server/src/main/java/com/cloud/agent/manager/allocator/impl/FirstFitAllocator.java:424
- Missing import for HashMap: add 'import java.util.HashMap;' so this compiles without errors.
Map<Long, Double> hostByComputedCapacity = new HashMap<>();
server/src/main/java/com/cloud/agent/manager/allocator/impl/FirstFitAllocator.java:419
- Missing import for ArrayList: please add 'import java.util.ArrayList;'.
return new Pair<>(new ArrayList<>(hostByComputedCapacity.keySet()), hostByComputedCapacity);
|
@blueorangutan package |
|
@sudo87 a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 14116 |
| return new Pair<>(new ArrayList<>(podsByCombinedCapacities.keySet()), podsByCombinedCapacities); | ||
| } | ||
|
|
||
| // order pods by combining cpu and memory capacity considering cpuToMemoeryWeight |
There was a problem hiding this comment.
should this be javadoc? (it seems the method name already is self-documenting)
There was a problem hiding this comment.
Sure @DaanHoogland, will remove comment in next commit.
|
@blueorangutan test |
|
@rosi-shapeblue a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests |
|
[SF] Trillian test result (tid-13752)
|
|
@blueorangutan test |
|
@sudo87 a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests |
RosiKyu
left a comment
RosiKyu
left a comment
There was a problem hiding this comment.
LGTM. Verification passed.
|
[SF] Trillian test result (tid-13764)
|
|
This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch. |
github-project-automation
bot
moved this from In Progress
to Done
in Apache CloudStack 4.21.0
* CPU to Memory weight based algorithm to order cluster host.capacityType.to.order.clusters config will support new algorithm: COMBINED which will work with host.capacityType.to.order.clusters.cputomemoryweight and capacity will be computed based on CPU and memory both and using weight factor * minor changes * add unit tests * update desc and add validation * handle copilot review comments * add log indicating chosen capacityType for ordering --------- Co-authored-by: Rohit Yadav <rohit.yadav@shapeblue.com>
7 participants
Description
This PR introduces new value: "COMBINED" for config: "host.capacityType.to.order.clusters", which will be used to order cluster, host and pods based on CPU and Memory both.
COMBINED will work with "host.capacityType.to.order.clusters.cputomemoryweight" and overall capacity for cluster/pod/host will be computed based on CPU and memory using weight factor.
The allocator will need to first calculate the combined allocation/usage metric (as follows), before sorting the clusters/pods/hosts to return a ordered list of hosts by this metric, for example:
For each host, define metric as:
Doc PR: apache/cloudstack-documentation#524
Types of changes
Feature/Enhancement Scale or Bug Severity
Feature/Enhancement Scale
Bug Severity
Screenshots (if appropriate):
How Has This Been Tested?
How did you try to break this feature and the system with this change?