Exam 5V0-21.21 topic 1 question 49 discussion

Actual exam question from VMware's 5V0-21.21

Question #: 49
Topic #: 1

An architect is designing a new Hybrid five node vSAN cluster for a customer who is planning to use vRealize Automation to provision 600 virtual machines into that cluster, with expected growth up to 1,000 VM. Each VM has a 40GB thick-provisioned disk.
Which flash disk size is required for the cache tier per ESXi node to meet all requirements?

A. 400GB
B. 700GB
C. 600GB
D. 800GB

Show Suggested Answer

Suggested Answer: D 🗳️

by wabim71208 at Oct. 8, 2021, 9:40 a.m.

Comments

Submit Cancel

wabim71208

Highly Voted 3 years, 1 month ago

40GB x 1000VMs = 40 000 GB total max expected usage 40 000GB / 5 node cluster = 8000 GB per node cache should be 10% so 800GB SSD is needed. However this question is stupid from real life perspective. one host can run up to 200 VMs - so what with MM, any failure ? Cluster should be sized with 6 ESXi hosts minumim Also cache should be 10% of raw HDD storage - before RAID policies applied. So what RAID 0? What architect is mentioned in this question? I belive learning one :D :D

upvoted 5 times

...

ranauro

Most Recent 1 year, 4 months ago

Selected Answer: C

in vsan 7 you can only use disk with 600gb max size

upvoted 1 times

...

jsi928

1 year, 4 months ago

C. The Cache disk size is capped at 600GB in 7.0. Even if you use a bigger disk, only 600GB of it will be used. https://kb.vmware.com/s/article/89485

upvoted 2 times

...

EvilRob

1 year, 6 months ago

Doesn't change the answer to this question but in 7.0, the usable Cache disk size is 600GB, increased to 1.6TB in 8.0 https://kb.vmware.com/s/article/89485

upvoted 3 times

...

Lazylinux

2 years, 11 months ago

D for sure cache 10% of total node capacity and hence 8000GB = 800GD SSD cache

upvoted 2 times

kernelkraut

8 months, 2 weeks ago

I would agree, you should aim for 10% caching minimum. However, there is the 600GB maximum to consider. There is an option to go beyond that with v8, at additional overhead (RAM per node). As they expect to grow, I would install 800GB disks with only 600 usable. Once they grow beyond the 600 machines, enable the option to go beyond that and accept the additional overhead.

upvoted 1 times

...

Exam 5V0-21.21 All Questions

View all questions & answers for the 5V0-21.21 exam