Exit
  • Global community
    • Language:
      • Deutsch
      • English
      • Español
      • Français
      • Português
  • 日本語コミュニティ
  • 한국 커뮤니티
0

Optimal AWS EC2 Instance Type to run Indesign Server?

Community Beginner ,
Sep 08, 2023 Sep 08, 2023

Hallo everyone,
I'm new to developing for indesign server and I'm tasked to figure out which instance type gives us the best performance. If anyone has experience with running Indesign Server in AWS I'd like to hear about your experience. I'd also be interested in any option we could configure to improve performance.

TOPICS
Experiment , Performance
452
Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
Sep 11, 2023 Sep 11, 2023

Small update for anyone who stumples upon this thread:

- There seems to be no global config, only the joboptions for pdf creation

- Single core performance is the most important for Indesign Server. I have only tested with t3.xlarge and R6i.large until now but tendency is clear. Maybe mac might be the best option, if the thing about graphics acceleration wasn't a misunderstanding ( further tests required )

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Mentor ,
Sep 11, 2023 Sep 11, 2023

Have you tried multiple InDesign Server instances in parallel on the same hardware, or do you have to optimize for a single job at a time?

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
Sep 12, 2023 Sep 12, 2023
LATEST

I only tested with a single Indesign Server instance. I used a sample script + template etc. from another project and used K6 for a small load test with 1-5 virtual users. For the actual implementation we probably want to use one Intance per vCPU + autoscaling, even if this will be a bit expensive in terms of licensing. 

For the current objective we just try to make requests go as fast as we can and then decide how we handle scaling without breaking the bank 😉

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines