Tools to help with the upload of the LOFAR ELAIS-N1 public data to the S3 service of the Amazon Cloud.
There are 20 datasets at the moment corresponding to observations of both cycle 0 and 2. The list of files is stored in the directory lists
.
After staging the data and creating the voms proxy certificate, the list of input URLs is entered to the command check_S3.py
. It filters the input list and outputs the files that are not yet in S3.
The list of URLs can be entered to download.py
to download the data and store it in S3. This can be done in parallel using the command xargs
.
Example:
python check_S3.py retrieval/003/surls_successful.txt | tee surls_no_003.txt
cat surls_no_003.txt | xargs -P 36 -n 1 python download.py