Good server consolidation strategy
Consolidation of VMs is essential to achieve energy optimization in cloud
environments such as OpenStack. As VMs are spinned up and/or moved over time,
it becomes necessary to migrate VMs among servers to lower the costs. However,
migration of VMs introduces runtime overheads and consumes extra energy, thus
a good server consolidation strategy should carefully plan for migration in
order to both minimize energy consumption and comply to the various SLAs.
-
class watcher.decision_engine.strategy.strategies.basic_consolidation.BasicConsolidation(config, osc=None)[source]
Bases: watcher.decision_engine.strategy.strategies.base.ServerConsolidationBaseStrategy
Basic offline consolidation using live migration
Description
This is server consolidation algorithm which not only minimizes the overall
number of used servers, but also minimizes the number of migrations.
Requirements
- You must have at least 2 physical compute nodes to run this strategy.
Limitations
- It has been developed only for tests.
- It assumes that the virtual machine and the compute node are on the same
private network.
- It assumes that live migrations are possible.
Spec URL
<None>
-
add_change_service_state(resource_id, state)[source]
-
add_migration(resource_id, migration_type, source_node, destination_node)[source]
-
calculate_migration_efficacy()[source]
Calculate migration efficacy
| Returns: | The efficacy tells us that every VM migration resulted
in releasing on node |
|---|
-
calculate_num_migrations(sorted_instances, node_to_release, sorted_score)[source]
-
calculate_score_instance(instance)[source]
Calculate Score of virtual machine
| Parameters: | instance – the virtual machine |
|---|
| Returns: | score |
|---|
-
calculate_score_node(node)[source]
Calculate the score that represent the utilization level
| Parameters: | node – ComputeNode instance |
|---|
| Returns: | Score for the given compute node |
|---|
| Return type: | float |
|---|
-
calculate_weight(compute_resource, total_cores_used, total_disk_used, total_memory_used)[source]
Calculate weight of every resource
| Parameters: |
- compute_resource –
- total_cores_used –
- total_disk_used –
- total_memory_used –
|
|---|
| Returns: |
|
|---|
-
check_migration(source_node, destination_node, instance_to_migrate)[source]
Check if the migration is possible
| Parameters: |
- source_node – the current node of the virtual machine
- destination_node – the destination of the virtual machine
- instance_to_migrate – the instance / virtual machine
|
|---|
| Returns: | True if the there is enough place otherwise false
|
|---|
-
check_threshold(destination_node, total_cores, total_disk, total_mem)[source]
Check threshold
Check the threshold value defined by the ratio of
aggregated CPU capacity of VMs on one node to CPU capacity
of this node must not exceed the threshold value.
| Parameters: |
- destination_node – the destination of the virtual machine
- total_cores – total cores of the virtual machine
- total_disk – total disk size used by the virtual machine
- total_mem – total memory used by the virtual machine
|
|---|
| Returns: | True if the threshold is not exceed
|
|---|
-
compute_score_of_nodes()[source]
Calculate score of nodes based on load by VMs
-
create_migration_instance(mig_instance, mig_source_node, mig_destination_node)[source]
Create migration VM
-
do_execute()[source]
-
classmethod get_display_name()[source]
-
classmethod get_name()[source]
-
classmethod get_schema()[source]
-
classmethod get_translatable_display_name()[source]
-
node_and_instance_score(sorted_scores)[source]
Get List of VMs from node
-
post_execute()[source]
-
pre_execute()[source]
-
unsuccessful_migration_actualization(number_migrations, unsuccessful_migration)[source]