You need to reduce the cost of virtual machines (VM) for your organization. After reviewing different options, you decide to leverage preemptible VM instances. Which application is suitable for preemptible VMs?
A.
A scalable in-memory caching system.
B.
The organization's public-facing website.
C.
A distributed, eventually consistent NoSQL database cluster with sufficient quorum.
D.
A GPU-accelerated video rendering platform that retrieves and stores videos in a storage bucket.
A
A GPU-accelerated video rendering platform that retrieves and stores videos in a storage bucket: Video rendering requires a stable and powerful infrastructure with persistent storage, which is not provided by preemptible VMs. Additionally, GPUs are not available on all preemptible VM instances.
D GPU-accelerated video rendering platform that retrieves and stores videos in a storage bucket: Video rendering requires a stable and powerful infrastructure with persistent storage, which is not provided by preemptible VMs. Additionally, GPUs are not available on all preemptible VM instances.
It seems A & D both as suitable answer. But I'll go with A, as attaching GPU with preemptible VM will increase cost, and in this question the purpose of opting for preemptible VM is reducing cost.
https://cloud.google.com/compute/docs/instances/preemptible#preemptible-with-gpu
ans: A
Preemptible VMs are best suited for fault-tolerant, non-critical applications due to their temporary nature. Among the options listed, A, a scalable in-memory caching system, aligns well with preemptible instances as it can handle interruptions and doesn't require continuous uptime.
Ans is D
Video rendering service is like application type called Batch job. Therefore, we can use instance type preemptible for them. If they complete task, they could be destroy and generate new instance to work continuously next task.
Why not A? "A scalable in-memory caching system."
In general a caching system is not critical to the function of an application.
If the cache is down it will cause requests to have cache miss and query the DB instead. User requests will still get served albeit slower.
In addition the answer specifies that the caching system is "scalable" reducing further the impact of 1 VM getting preempted, ie traffic can be automatically redirected to other cache replicas.
To me all other answers seem to have a more severe impact on the user in case the VM is preempted.
i think is C, database cluster is storage, and distributed, eventually consistence is resistant for the preempted. and sufficient quorum can ensure the DB transitions.
A voting comment increases the vote count for the chosen answer by one.
Upvoting a comment with a selected answer will also increase the vote count towards that answer by one.
So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.
Charun
Highly Voted 3 years, 5 months agoNiveusSol
1 year, 7 months agoxarom
1 year, 3 months agodevopsbatch
Highly Voted 3 years, 5 months agoakg001
3 years, 5 months agoNiveusSol
1 year, 7 months ago6a8c7ad
Most Recent 3 months, 2 weeks agowinston9
5 months, 2 weeks agojinaldesailive
8 months, 3 weeks agoFI22
8 months, 3 weeks agoMFay
11 months, 1 week agojomonkp
12 months agoJason_Cloud_at
1 year, 1 month agoWatcharin_start
1 year, 6 months agoizekc
1 year, 8 months agoeks4x
1 year, 11 months agokisshs
1 year, 11 months agohanweiCN
1 year, 12 months agoeliC
2 years, 5 months agopsyx21
2 years, 6 months agoTNT87
3 years, 2 months ago