Issue
In sync mode working in SGE, secondary analysis creates new job and submits to cluster. However this creates a blocking situation if the available open slots are exceeded by existing alignment jobs. Under this situation, the alignment jobs waits on secondaryanalysis (cufflinks or featurecount) that are in qw form.
For example:
Start of job
available slots = 10
samples = 10
used slots = 10
Completion of alignment, Start of secondary analysis
Issue
In sync mode working in SGE, secondary analysis creates new job and submits to cluster. However this creates a blocking situation if the available open slots are exceeded by existing alignment jobs. Under this situation, the alignment jobs waits on secondaryanalysis (cufflinks or featurecount) that are in qw form.
For example:
@kamyshova Do you have some input on possible solutions? I am still in favor of solution proposed here: https://github.com/epam/fonda/issues/162
If not, what do you think of possible solutions?
Thanks