MultiThread and async mode.

GemBox.Document
1

We are experiencing severe performance issues in our web-based application after upgrading to your latest library version.

We use your component to convert Docxs, Images and HTML content. The issues began after we updated our system code and included your updated library.

We urgently need clarification on how your library behaves under load, especially in an ASP.NET environment. Could you please help answer the following:

  1. Does your DOCX to PDF (and PDF-to-DOCX) conversion use synchronous or asynchronous processing internally?
  2. Does the conversion block the calling thread (e.g., in a web request)?
  3. Is your library thread-safe when used with multiple concurrent requests?
  4. Does it lock any resources (files, memory, handles, etc.) that could cause contention under load?
  5. Do you recommend a limit to how many conversions should run concurrently in production?
  6. Is there any way to enable internal logging of conversion time, thread usage, or similar for troubleshooting?

This is critical for us to understand as we are diagnosing thread starvation and performance bottlenecks.

Thank you in advance for your urgent assistance.

2

Hi,

We are experiencing severe performance issues in our web-based application after upgrading to your latest library version.

What kind of issues? Can you tell us a bit more about how you’re observing the problem, are you getting exceptions?
If you revert to an older version, does it avoid the problem?

Regarding your questions:

  1. All actions are done synchronously.
  2. Yes.
  3. Yes, GemBox.Document is thread-safe in general. However, it is not safe to manipulate with the same DocumentModel object on multiple threads.
  4. There should be no locking.
  5. It will depend on your resources and the size of the documents you’re currently converting.
  6. GemBox.Document doesn’t have such a feature; however, you could add some profiling to your thread.

This is critical for us to understand as we are diagnosing thread starvation and performance bottlenecks.

Is the problem happening with a specific input Word documents or with any document?
Is the problem happening with input PDF documents, not with Word documents?
Are you able to reproduce the issue locally?