Error: Cannot construct DataBuffer

#19

I was with the same problem when I want to run “S-1 SLC to GRD”. After several hours to several processes and attempts I got these two procedures:
-Update Java to the latest version from the official website.
My notebook:
i72620m 2.28ghz
8gb ram
Windows 10

Image Sentinel 1:
S1A_IW_GRDH_1SDV_20160114T092037_20160114T092112_009488_00DC51_839E (9.23gb)

SNAP/options/

Pasted image384×989 48.4 KB

***these were the three things I did time; but perhaps only one of them the problem is solved.

I hope this is helpful…

1 Like
#20

Updating java is probably not one of the helpful things. SNAP come with its own java jre and this is not updated by the normal update process, or have you changed the jre folder in the installation directory?

Most helpful is probably to use the file cache option.

#21

Hi everyone

I am having the same problem with “DataBuffer” and i’ve been reading the comments in the forum. So what is the solution to our problem in the end? My computer is using 4 GB RAM and i really need to use SNAP for S-1 SAR data.

#22

Have you already activated the file cache option?

#23

I do no not change, just i go to java web page and download de version in screen, however do you try with the other two things?

#24

see the images attached, i try in 2 pc’s with de same problem.

#25

Hi again

I have tried to do the same processing in a computer with 32 GB ram and it works perfectly.

#26

Is increasing the RAM the solution to this issue ?

#27

It worked immediately when i did it in a super computer with 32 GB RAM. My laptop and the desktop i am working with usually have 4GB RAM and i could never do anything with SNAP or S-1 toolbox. Now I did every process i wanted so i guess it is a RAM issue.

#28

Long graphs also consume more memory so breaking them into pieces may help.

2 Likes
#29

This problem is very frustrating. From what I’ve read I can see it is a RAM issue. My machine has 4GB of RAM. Damn!

1 Like
#30

If you computer cannot handle processing full products and large graphs,

  1. Create subsets of smaller areas of interest
  2. Use smaller graphs or single operators
  3. In the tools options s1tbx dialog use the file cache for sentinel-1 reading
  4. In the tools options performance dialog use smaller tiles
  5. Close other applications using memory
2 Likes
Cannot set heapsize on ubuntu 16.04
#31

I was facing same problem during last weak, when i was try to calibrate SLC product of Sentinel-1A. Then I tried to create space from hard drive and that worked :). Actually I tried same process in my university P.C, and that went successful in first attempt. The output image was too heavy, having size of around 20 GB. Then, I created the space of more than 20 GB at my laptop also, which worked and I didn’t get error regarding data buffer construction.

1 Like
#32

Does optimizing through the GUI impact performance & processing when implementing using snappy/GPF? I’m trying to do a processing graph that takes less than a minute in the GUI, but when I’ve coded it out using snappy and GPF, it fails with the DataBuffer error or JavaHeapSpace.

#33

Maybe it is necessary to edit the gpt.vmoptions file.
Uncomment the line with the xmx option and set it to e.g. -Xmx6G

#34

I found that using the GUI optimization did nothing for processing through python\snappy, and only improves performance when doing processing within the SNAP GUI.

To solve my issue, it came down to modifying the snappy.ini file as has previously been posted.

1 Like
#35

<img src="/uploads/default/original/2X/7/7817b0bc672889f58258bac4234664f9e1f96e67.jpg" width=“690” height="3
Hello everyone!
I have a problem with Sentinel-1 in Steps 1 (Calibrate).

#36

I don’t have a solution but I fixed the inserted image for you.

blob.jpg1366×768 374 KB

#37

Hi. Thank you so much!

#38

Hello,

I had the same problem using gpt.exe - using GUI it worked fine. 8BG ram here, but my gpt proces is using only about 2,2GB

The solution which helped me:

Thanks a lot.