Running instance when not using Nimble

0

Hi,

I am testing the solution (mainly the remote workstation for now). I have created a new AWS account just for that. Right now I am not using Nimble, I see 2 instances running all the time:

1x c5.large = RCS Capacity
1x m5.4xlarge = Worker Fleet

1x t3.medium = activedirectory is Stopped (good) and of course the nimble studio builder is stopped (will terminate this instance eventually). Looking at the billing, looks like I will be charge a good amount even if I am not using the solution. I configured the fleet instance as on-demand. Is this normal, because I am testing nimble right now, should I always delete the studio when not using it, but what about in production - what are the best practices for optimizing cost? I would not mind have AWS credits for testing nimble, just saying...

eeight
asked 3 years ago289 views
6 Answers
0

Hello,

There are a handful of services needed to keep a Nimble Studio instance active. We are currently working on our pricing documentation to better explain the needed costs to keep Nimble Studio running. The instances you point out are part of the resources needed to deploy an active production studio. The RCS instance is the Deadline (used for rendering) render server and is kept running so you can submit render jobs to it. While it is running when you are not using it, it is needed to make sure when you do submit something, the render farm is ready to distribute your jobs to the farm.

corban-aws

AWS
answered 3 years ago
0

Hi Corban,

I just received the first invoice, a bit hard to swallow I have to admit, because I have a feeling that I am doing QA on nimble studio. Also I was under the impression that I could get either AWS credits OR at least one month free (from http://www.cgchannel.com/2021/04/aws-launches-amazon-nimble-studio/ | At the time of writing, new users of Amazon Nimble studio get a one-month free trial). How to apply?

In my case, the 1x m5.4xlarge = Worker Fleet was always on, I had to manually click on "Stop instance".

Why not having a script launching the c5.large when requesting a render instead of having it always on waiting for someone to use the render farm?

Edited by: eeight on Jun 10, 2021 7:33 AM

eeight
answered 3 years ago
0

Hello,

I reached out via PM.

aws-corban

AWS
answered 3 years ago
0

Stopping the WorkerFleet instance was not enough. I read that "The number of render workers are only used when you submit a job to the farm." but in the end, I had to modify the auto scaling groups for the WorkerFleet:
Desired capacity & Minimum capacity were set to 1 (now 0)

Maybe something with this step in the studio builder script:
2- Enter the Min number of render workers. (This will be how many are initally deployed)
^ we cannot use 0, we have to enter 1.

Thanks

Edited by: eeight on Jun 21, 2021 9:00 AM

eeight
answered 3 years ago
0

I'm also doing some testing of Nimble studio for the first month but am being charged. Do you have to apply for this free month?
It would be great to get a breakdown of where a studio will expect to be charged so we can fully evaluate the cost of Nimble Studio. Is there any documentation on the default network structure that's set up once you build the studio? Is seems there's a lot of moving parts that get created when you set up a studio. I'm going through trying to figure out what's going on in each subnet, which security groups/nacls are applied, what each ec2 instance is doing etc. Is there any documentation on what gets created by default?

answered 3 years ago
0

Nimble Studio does not have free tier. As for updated pricing, this link should help: https://aws.amazon.com/nimble-studio/pricing/

While we don't have an architecture document that covers all aspects of Nimble Studio we are continuously updating our documentation so I would encourage you to check back. This is a good suggestion and I have shared it internally with out documentation team.

Thanks for using Nimble Studio,

gunnar-aws

AWS
answered 3 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions