OverCommit CPU and Memory

This feature  is implemented for following hypervisor  
  1-VMWARE
  2-KVM 
  3-XEN

1

UI/API check

1-Add cluster using  API addCluster

add cluster API should have new parameters  cpuovercommitratio and memoryovercommitratio

P0

Y

Pass

2

UI/API check

1- update overcommit ratio using API updateCluster with parameter cpuovercommitratio,memoryovercommitratio.

response should contain " details of the updated cluster along with the overcommit ratios"

P0

Y

Pass

3

DB check

1-Logoin through UI
2-create a cluster with overcommit factor x
3-Login to DB
4-check cluster_detail table

1-There should be   attribute ,name,value for each cluster .

P0

Y

Pass

 Deploy VM in cluster

4

Deploy vm in a cluster with ram overcommit factor 2x

1-deploy vm with ram= x
2- deploy another VM with ram<=x

1- VM in step 1 should get deployed
2-VM in step 2 should also get deployed

P0

Functional

Y

Y

Y

Y

N

N

5

Deploy VM in cluster with ram overcommit factor 2x

1-deploy vm with ram= x
2- deploy another VM with ram>x

1- VM in step 1 should get deployed
2-VM in step 2 should not  get deployed

P0

Functional

Y

Y

Y

Y

N

N

6

Deploy vm in a cluster with cpu overcommit factor 2x

1-deploy vm with cpu= x
2- deploy another VM with cpu<=x

1- VM in step 1 should get deployed
2-VM in step 2 should also get deployed

P0

Functional

Y

Y

Y

Y

N

N

7

Deploy vm in a cluster with cpu overcommit factor 2x

1-deploy vm with cpu= x
2- deploy another VM with cpu >x

1- VM in step 1 should get deployed
2-VM in step 2 should not  get deployed

P0

Functional

Y

Y

Y

Y

N

N

8

Deploy vm in cluster with overcommit factor x . and total available ram = M

1-Deploy vm with aSO where ram=y GB
2-max and min will be y GB and y/x GB
3-Deploy  vms  such sum of all y>M

1-CS will choose a value in y/x to y such that total required ram will be < M.

P0

Functional(hold)

Y

Y

Y

Y

N

N

9

Addition of host to cluster

1- add host which have overcommit capacity
2-Add another host wich don't have overcommit capacity

1-CS should allow addition of host independent of overcommit capacity

P1

Fuctional

Y

Y

Y

Y

N

N

10

Change of Overcommit factor

1-while creating cluster set overcommit factor to x
2-deploy few vms and add soome host
3-Change overcommit factor to y

1-it show not affect existing vms(until unless they are stop/start)
2-no effect host (assumption of homogeneous host)

P1

Functional

Y

Y

Y

Y

N

N

"overcommit factor update " effect on vm

1-Login Through UI
2-Create a cluster with overcommit ratio x
3-Deploy a vm
4-Change overcommit ratio to y

11

5-when y<x and there is no(less than required) resource left in cluster according to new overcommit ratio
6-Deploy a  vm.

1-VM deployed in step 3 should not have any effect of y 
2-step 6  should fail  for both CPU and RAM

P0

Functional

Y

Y

Y

Y

N

N

12

5-when y<x and there is enough resource left in cluster according to new overcommit ratio
6-Deploy a  vm.

1-VM deployed in step 3 should not have any effect of y 
2-step 6  should be successful for both CPU and RAM

P0

Functional

Y

Y

Y

Y

N

N

13

5-when y>x and there is enough resource left in cluster according to new overcommit ratio
6-Deploy a  vm.

1-VM deployed in step 3 should not have any effect of y
2-step 6  should be successful for both CPU and RAM

P0

Functional

Y

Y

Y

Y

N

N

14

5-when y>x and there is not enouh resource left in cluster according to new overcommit ratio
6-Deploy a  vm.

1-VM deployed in step 3 should not have any effect of y
2-step 6  should fail for both CPU and RAM

P0

Functional

Y

Y

Y

Y

N

N

 VM Life cycle

15

Reboot

1-set cluster overcommit ratio x
2-Deploy some vm such that no resources are available on current host
3-change overcommit ratio to y

16

1-Reboot all the vms in deployed in step 2 when x>y

all vm should come up without any failure

P1

Functional

Y

17

1-Reboot all the vms in deployed in step 2 when x<y

all vm should come up without any failure

P1

Functional

Y

18

stop/start(destroy/restore)

1-Stop and start all the vms in deployed in step 2 when x>y

1-all vm should get resource depend on new overcommit ratio
2-in case no resource available some vm will fail to start

P1

Functional

Y

19

1-Stop and start all the vms in deployed in step 2 when x<y

P1

Functional

Y

20

1-Stop and start one by one

some vm will not start

P1

Functional

21

(manage/unmanage-Disable/enable  cluster)

1-Create a cluster with overcommit ratio x
2-unmanage-manage cluster
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="258d6bc5-990b-4061-97c1-0c0be1e78c43"><ac:plain-text-body><![CDATA[

1-Overcommit ratio should not change

P1

Functional

Y

Y

Y

Y

N

N

Fail

[https://issues.apache.org/jira/browse/CLOUDSTACK-1534
]]></ac:plain-text-body></ac:structured-macro>
]

22

upgrade path

Pre upgrade steps
1-Deploy  some vms
2-Have some snapshot and template

Post upgrade steps
1-add cluster with overcommit factor x
2-Deploy a vm  where required resource for vm <= resource available in cluster according to overcommit factor
3-Deploy a vm where required resource for vm > resource available in cluster according to overcommit factor

1- Upgrade should be successful .
2-Check cluster detail table it should have some attribute to store overcommit factor for cluster.
3-in postupgrade step-2 vm should get deployed successfully .
4- In postupgrade step-3 vm deployment should fail. 

Upgrade

Y

Y

Y

Y

N

N

23

Number of max VM can be deployed (CPU overcommit factor x, actual cpu C) in case enough RAM  available.

1- Deploy a vm with a SO where cpu=y .

number of vm must be limited by n*y/x<C (nummber of vm *lower value of ram < total cpu in cluster)

P1

Functional

Y

Y

Y

Y

N

N

24

Number of max VM can be deployed (RAM overcommit factor x, actula ram M) in case enough CPUs are available.

1- Deploy a vm with a SO where ram=y ,limit(y/x ,y)

number of vm must be limited by n*y/x<M (nummber of vm *lower value of ram < total memory in cluster)

P1

Functional

Y

Y

Y

Y

N

N

25

Migrate vm from one cluster to another

vm should be able to migrate in any cluster ,having required resources available, irrespective of overcommit factor .

P0

Functional(hold)

Y

Y

Y

Y

N

N

26

Dash board display
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="1e1a3fc4-31be-46a6-9f60-8931450b2e33"><ac:plain-text-body><![CDATA[

1-Check dashboard display at cluster level

should display resources based on overcommit ratio

P1

Functional

Y

Y

Y

Y

N

N

[https://issues.apache.org/jira/browse/CLOUDSTACK-1538
]]></ac:plain-text-body></ac:structured-macro>
]

27

Change overcommit ratio

1-Create cluster with overcommit ratio x
2-have some vm
3-change overcommit ratio y(y<x)

1-Dash board display should not display >100% allocation

P3

Functional

Y

Y

Y

Y

N

N

28

overcommit ratio <1

1-Craete a cluster with overcommit ratio x
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="79e9a3be-3561-4d65-98c5-bafef8e09c78"><ac:plain-text-body><![CDATA[2-Change overcommit ratio to <1

1-Cs should not allow to set overcommit ratio <1

P3

Functional

Y

Y

Y

Y

N

N

Fail

[https://issues.apache.org/jira/browse/CLOUDSTACK-1528
]]></ac:plain-text-body></ac:structured-macro>

29

cluster.cpu.allocated.capacity.disablethreshold

set cluster.cpu.allocated.capacity.disablethreshold
to some value say .85

1-Should be calculated based on overcommit ratio

P2

Functional

Y

Y

Y

Y

N

N

30

cluster.cpu.allocated.capacity.notificationThreshold

set cluster.cpu.allocated.capacity.notificationThreshold
to some value

1-should be calculated based on overcommit ratio

P2

Functional

Y

Y

Y

Y

N

N

31

cpu.overpovisioning .factor

set cpu.overpovisioning .factor to some value

1-should be calculated based on overcommit ratio

P2

Functional

Y

Y

Y

Y

N

N