Response performance after upload via API

[alexanderadam]

We're currently struggling with high resource usage when processing a lot of documents.
Right now it's not possible to disable OCR processing even if we know that we don't need OCR in these cases or even the different document formats.

[jendib]

As you know the OCR is done by Tesseract which is done by another team/project so we cannot do much about the resource usage. From my experience a fast CPU is indeed required.