Hello there, I’m using Java GroupDocs.Conversion, so I’d like to know, what is the limit for a file large enough to be unsupported?
Thanks in advance
There is no limit. However, in case of any such issue, do let us know and share the problematic file.
Hi Atir, in a java groupdocs library what is the maximum file size you have tested with and what is the memory and CPU configuration that you have used. We have a requirement to support around 2GB for our customers
Hello @amitdash ,
Could you please clarify which file format you are planning to convert and what output format you expect? We believe it may be challenging to test files of around 2 GB in size.
In general, we recommend the following baseline configuration:
- CPU: 2–4 cores
- RAM: 2–4 GB
- JVM heap: -Xms512m -Xmx2g
However, based on your scenario, you would likely need at least:
- CPU: 4–8 cores
- RAM: 4–8+ GB
- JVM heap: -Xmx4g or -Xmx8g
Please note that these are approximate recommendations, as performance depends on several factors, including the file format, content complexity (fonts, graphics, tables, formulas), and the type of conversion (for example, DOCX → PDF is more resource-intensive than TXT → PDF).
If you could share a sample document for testing, we would be able to provide more accurate recommendations tailored to your case.
hi Even, we are planning to convert docx into pdf
Hello @amitdash ,
We think you can certainly try following our recommendations, but most likely even 8 GB of RAM will not be enough for such large documents.
As mentioned earlier, we could perform an investigation on our side using your file to provide you with the most accurate recommendations. You can share a link via Google Drive for this large document (my email for access is evgen.efimov@aspose.com ) or any other cloud storage that is convenient for you.
Here is the docx : https://drive.google.com/drive/folders/1F4jfWu8NaC769fkiXg5YcdoCIkaW7vrB
I gave you access.
Could you please perform test using this file.
When we tried on our end. We used 28GB memory and 2CPU Pod for processing but we were hitting : OutOfMemoryError: Requested array size exceeds VM limit
java.lang.OutOfMemoryError: Requested array size exceeds VM limit
Please check on your end and suggest way forward.
Thanks !
Hello @anmolgarg1134 and @amitdash ,
Thank you for sharing the document. I have forwarded it to our development team for investigation and created a corresponding ticket in our tracking system with ID CONVERSIONJAVA-3131. As soon as I receive feedback from them, I will inform you.
Hello @anmolgarg1134 ,
Unfortunately, we do not have any news regarding this issue at the moment. Tomorrow we will check with the development team about the progress of their investigation and will contact you as soon as we have an update.
I investigated the issue and found that the root cause is the limitation of Java array capacity. Currently, byte[] arrays are used for storing document data in GroupDocs.Conversion and its underlying dependencies.
Due to Java platform limitations, processing very large files (around 2+ GB) may lead to java.lang.OutOfMemoryError: Requested array size exceeds VM limit, even when sufficient heap memory is allocated. This is because Java arrays are indexed by int and therefore have a practical size limit of approximately 2 GB.
To support files larger than ~2 GB, significant architectural changes would be required in GroupDocs.Conversion and the underlying libraries, including moving away from loading entire document content into contiguous byte arrays and adopting a streaming/chunked processing approach.
We will further analyze the feasibility and estimated effort required for such an enhancement and evaluate the possibility of including it in a future roadmap.