Azure AI Translator Container Memory Leak Symptoms Causes and Solutions Inquiry

We configured AI Translator Container on 4 servers a month ago (June 2025).

There is no problem with the translation itself, but we have confirmed that there is a memory leak as shown in the image below.

I would like to ask how to identify the cause of the memory leak and resolve it.

• How to configure
  • Install Docker on each of the 4 servers and install the AI Translator container image on them.
• Pages referenced during configuration
  • https://learn.microsoft.com/en-us/azure/ai-services/translator/containers/install-run?tabs=connected
  • https://learn.microsoft.com/en-us/azure/ai-services/translator/containers/configuration
• Spec.
  • Server OS : UBUNTU 22.04
  • CPU : 24 core
  • Memory : 48 GB
  • Azuer AI Translator container image version : 1.0.03051.790-amd64
  • Docker engine ver. 28.2.1
• Drive command docker run -d --rm -it -p 5000:5000 --memory 46g --cpus 23.9 \ --name cpaas_trans \ --log-driver=syslog \ --log-opt tag="container" \ -v /mnt/d/TranslatorContainer:/usr/local/models \ -e apikey=<<API_KEY>>\ -e eula=accept \ -e billing=<<BILLING_URL>> \ -e Languages=ko,en,ja,zh-Hans,zh-Hant,es,ru,vi,de,fr,it,pt,hu,pl \ mcr.microsoft.com/azure-cognitive-services/translator/text-translation:${TAG}
• Memory Leak graph (grafana)