Java memory issue with Simagis Live service
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Java memory issue with Simagis Live service
I’m using free Simagis Live service ( http://images.simagis.com/live/, beta ) to share my microscopy slides with collaborators. They say they you can upload Tiff files of any size but when I’m trying stream BigTIFF images over 1.5GB to my folder on server using Upload Utility, I’m getting Java error message “OurOfMemoryError Java heap space” . I have 3GB of memory on my machine but looks like Java does not let me use all of it. Does anybody know how to make Java use all the memory? Or how to upload large files to Simagis Live server bypassing memory issues?
-
rjlittlefield
- Site Admin
- Posts: 23626
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
ginaqwer, welcome aboard!
Java has a number of different memory limitations depending on which operating system, whether the program is stand-alone or running inside a browser, and how it is invoked. I have no experience with Simagis Live, so I have no idea which limit might apply. On 32-bit Windows, stand-alone Java (Sun's JRE) can get 1.6 GB at best, sometimes less. Java in a browser gets substantially less, and it varies by browser.
In any case, upload facilities usually stream the data through relatively small fixed size buffers, rather than trying to read the whole thing into memory and then trickle it up. The message you're getting suggests that either the Simagis Upload Utility is trying to read the whole thing, or that something else is going on to confuse the issue.
If Simagis provides some kind of support service or forum, it would be good to post your question there. If they are aware of the issue, then you should get a specific and reliable answer. If they are not aware of it, then you will at least have made them aware so they can address it in future development.
Switching subjects a little, I am curious about the combination of "microscopy slides" and 1.5 GB. What sort of setup are you using that can generate 1.5 GB of useful pixels in microscopy?
--Rik
Java has a number of different memory limitations depending on which operating system, whether the program is stand-alone or running inside a browser, and how it is invoked. I have no experience with Simagis Live, so I have no idea which limit might apply. On 32-bit Windows, stand-alone Java (Sun's JRE) can get 1.6 GB at best, sometimes less. Java in a browser gets substantially less, and it varies by browser.
In any case, upload facilities usually stream the data through relatively small fixed size buffers, rather than trying to read the whole thing into memory and then trickle it up. The message you're getting suggests that either the Simagis Upload Utility is trying to read the whole thing, or that something else is going on to confuse the issue.
If Simagis provides some kind of support service or forum, it would be good to post your question there. If they are aware of the issue, then you should get a specific and reliable answer. If they are not aware of it, then you will at least have made them aware so they can address it in future development.
Switching subjects a little, I am curious about the combination of "microscopy slides" and 1.5 GB. What sort of setup are you using that can generate 1.5 GB of useful pixels in microscopy?
--Rik
Thanks, Rik,
I have got message from support with settings for upload utility command line that forces Java to use more memory. It works now.
Here is example of large image, published http://images.simagis.com/live/public/4 ... 8c355b8e21
This is tissue slide at full resolution with on-line annotations. It was acquired as “mosaic” using microscope with automated stage.
Thanks for your help,
Gina
I have got message from support with settings for upload utility command line that forces Java to use more memory. It works now.
Here is example of large image, published http://images.simagis.com/live/public/4 ... 8c355b8e21
This is tissue slide at full resolution with on-line annotations. It was acquired as “mosaic” using microscope with automated stage.
Thanks for your help,
Gina