Which three things can be done to allow more virtual machines to be deployed into the cluster from this templa

A company has been utilizing templates in its environment. It is running a 10-node ESXi 5.x Cluster and DRS has not been configured. Several virtual machines have been deployed from this template and successfully powered on, but a newly deployed virtual machine will not power on. There appears to be adequate CPU and Memory resources available on the host. Which three things can be done to allow more virtual machines to be deployed into the cluster from this template? (Choose three.)

A.
Deploy the virtual machine to a different host using the same datastore

B.
Enable DRS on the cluster to balance the virtual load out across hosts

C.
Increase the virtual machine memory reservation

D.
Move the swap file to a different location

E.
Select a different datastore for the virtual machine

Explanation:
(Reviewed)
By increasing the VM memory reservation, the swap file will be smaller. See the VMware KBhttp://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1003755

Quote from KB:
If you make reservations for your virtual machine’s that are equal to the amount of RAM assigned to them, swapping and page sharing does not occur. You can over commit pretty heavily if you are comfortable with poorer performance. If you do not set reservations, ESX host creates a .vswp file equal to the difference between the amount of physical memory assigned to the virtual machine and the reservation it has. By default, memory reservations are set to 0. If you have a virtual machine with 2GB of memory without a reservation, it creates a 2GB .vswp file when it is powered on. The virtual machine starts using the .vswp file if the server is out of physical RAM. If you set a 1GB reservation, it creates a 1GB .vswp file. The .vswp files are what allows for memory overcommitment.

8 Comments on “Which three things can be done to allow more virtual machines to be deployed into the cluster from this templa

  1. Scott says:

    I believe its BDE…

    the question asks.. Which three things can be done to allow more virtual machines to be deployed into the cluster from this template? (Choose three.)

    A is wrong, since they clearly indicate “There appears to be adequate CPU and Memory resources available on the host” so we know the issue is disk. that is why using a different host and THIS overcommitted disk will not resolve the issue.

    B is correct, because with drs, they cluster will balance itself better than was done manually. drs will either make recommendations or perform migrations automatically and make changes to free up space on various datastores.

    C isn’t correct. it would only solve the problem for 1 vm and the questions is how to deploy more vms using the template.

    D is correct as moving the swapfile to another datastore will help relieve space on the datastore and could be considerably if no reservations are set. 4gb to 8GB easily.

    E is correct as this option would allow more vms to be deployed.

  2. Leandro Gomes says:

    BDE is wrong too:

    Question: A company has been utilizing templates in its environment. It is running a 10-node ESXi 5.x Cluster and DRS has not been configured. Several virtual machines have been deployed from this template and successfully powered on, but a newly deployed virtual machine will not power on. There appears to be adequate CPU and Memory resources available on the host.

    Which three things can be done to allow more virtual machines to be deployed into the cluster from this template? (Choose three.)
    Incorrect response(s):
    Enable DRS on the cluster to balance the virtual load out across hosts
    Move the swap file to a different location
    Select a different datastore for the virtual machine
    Note: This response is either incorrect, or additional responses may be required to receive a 100% score for this question.

  3. Andrew Dawes says:

    This is likely CDE because memory reservations reduce the size of the swap file for the VM in question. Since storage on the LUN is likely the cause, we would indicate for things that would help with storage. DRS will not be it because they do not state explicitly “STORAGE DRS”, therefore this will not fix it as memory and CPU are fine. Previous posts indicate that moving the swap file will likely help, and I agree. Obviously a different datastore will likely alleviate this issue. Thus, I kind of disagree with most of the comments, though I can be completely wrong about that. I say that “C” is right as the swap file is equal to VM memory minus reservation. If all memory for a VM is also reserved for that VM, the VM’s swap size is zero. Thus, we can assume the storage improvement will be helpful in this problem.

  4. Mark says:

    I agree with CDE. Having concluded that the problem is indeed storage-related, we have to consider how we can free up storage.

    Placing the VM on a different datastore is obvious. So E is a given. By extension, so is D.
    Increasing the vRAM reservation will create a smaller vswp-file and will therefor work. (Provided that – as is stated in the question – the machine is OFF. If it’s running, the vswp-file won’t be decreased until the VM has been turned off). That makes C an answer.

    A, deploying on the same datastore makes no sense. As for B, just remember IT exam writers have no life and like to mess with the candidates by playing with words. DRS is not SDRS.

    Still, what are the odds that you have enough storage to deploy a VM, but not enough to accommodate the vswp-file upon turning it on? One lucky admin.


Leave a Reply

Your email address will not be published. Required fields are marked *