You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Tapis allows us to use macro variables to define what the job's execution folder should be, etc. Some of these macro variables include the job's uuid (the default) and the job's name.
Tapis's default job working directory currently uses Job UUID to create job folders, while is safe and effective from machine point of view. However, we plan to implement job folder names based on user supplied string. This way a user can browse their job outputs intutively on a filesystem instead of having to map Job UUID with corresponding folder.
This is already feasible to implement to certain extent using the files API (checking if the folder with the same name already exists before making the submitJob API call, etc.), however the number of API round trips needed for this can be numerous in edge cases.
So, we would like to propose adding a new macro JOB_FOLDER_NAME_AUTOINCREMENT (the name can be different) that has following behaviors
Create's a job folder with a given name say "myjob", and if this folder already exists it will create a new job folder with name as "myjob_1". Similarity when "myjob_1" exist, it will create a new folder as "myjob_2" and so on.
Would love to hear your thoughts on this, thanks
The text was updated successfully, but these errors were encountered:
Tapis allows us to use macro variables to define what the job's execution folder should be, etc. Some of these macro variables include the job's uuid (the default) and the job's name.
Tapis's default job working directory currently uses Job UUID to create job folders, while is safe and effective from machine point of view. However, we plan to implement job folder names based on user supplied string. This way a user can browse their job outputs intutively on a filesystem instead of having to map Job UUID with corresponding folder.
This is already feasible to implement to certain extent using the files API (checking if the folder with the same name already exists before making the submitJob API call, etc.), however the number of API round trips needed for this can be numerous in edge cases.
So, we would like to propose adding a new macro
JOB_FOLDER_NAME_AUTOINCREMENT
(the name can be different) that has following behaviorsCreate's a job folder with a given name say "myjob", and if this folder already exists it will create a new job folder with name as "myjob_1". Similarity when "myjob_1" exist, it will create a new folder as "myjob_2" and so on.
Would love to hear your thoughts on this, thanks
The text was updated successfully, but these errors were encountered: