Details
-
Type:
Story
-
Status: Done
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
-
Story Points:1
-
Epic Link:
-
Team:Data Facility
Description
Pending approval of RFC-578, please install the Goulding S18A bright star masks on the LSST cluster.
I recommend the following strategy:
- Rename the current /datasets/hsc/deepCoadd/BrightObjectMasks --> /datasets/hsc/deepCoadd/ArcturusMasks.
- Install the Goulding S18A bright star masks in /datasets/hsc/deepCoadd/GouldingMasksS18A.
- Create a soft link: /datasets/hsc/deepCoadd/BrightObjectMasks that points to /datasets/hsc/deepCoadd/GouldingMasksS18A.
This keeps both versions available, defaulting to the new masks. Users can explicitly select the old masks by creating a link in their rerun named deepCoadd/BrightObjectMasks that points to /datasets/hsc/deepCoadd/ArcturusMasks.
I am currently downloading the Goulding S18A masks to /scratch/pprice/GouldingMasksS18A.
Attachments
Issue Links
- is triggered by
-
- Implemented
Activity
Created IHS-1942 for the admin to unlock the data on GPFS.
Unless I hear objections, I'll proceed with the modified plan in my last comment.
Hsin-Fang Chiang
added a comment - Created IHS-1942 for the admin to unlock the data on GPFS.
Unless I hear objections, I'll proceed with the modified plan in my last comment. That's a better plan than my original suggestion; let's go with that!
Paul Price
added a comment - That's a better plan than my original suggestion; let's go with that! The code changes have merged, and the RFC has been adopted. Hsin-Fang Chiang, please proceed at your convenience. Thanks!
Paul Price
added a comment - The code changes have merged, and the RFC has been adopted. Hsin-Fang Chiang , please proceed at your convenience. Thanks! Paul Price I think I can safely ignore this file /scratch/pprice/GouldingMasksS18A/skyMap.pickle right?
Hsin-Fang Chiang
added a comment - Paul Price I think I can safely ignore this file /scratch/pprice/GouldingMasksS18A/skyMap.pickle right? Maybe. It's a copy of the usual RingsSkymap used for HSC. I used it for the partitioning, and so it may be useful in recording how the data were partitioned. But if you think that's standard enough, you're welcome to ignore it.
Paul Price
added a comment - Maybe. It's a copy of the usual RingsSkymap used for HSC. I used it for the partitioning, and so it may be useful in recording how the data were partitioned. But if you think that's standard enough, you're welcome to ignore it. Files are in place now and the new link is made.
Hsin-Fang Chiang
added a comment - Files are in place now and the new link is made. 
I want to propose a slightly different organizing.
I'd create a new folder at /datasets/hsc/BrightObjectMasks/ outside of the Butler repo, and contain ArcturusMasks and GouldingMasksS18A inside it (so they would be
/datasets/hsc/BrightObjectMasks/ArcturusMasks and /datasets/hsc/BrightObjectMasks/GouldingMasksS18A )
Then make /datasets/hsc/repo/deepCoadd/BrightObjectMasks a soft link to the /datasets/hsc/BrightObjectMasks/GouldingMasksS18A folder.
My intention is to do a little better with multiple versions (before we have better tools) and separate where we store both sets of data versus the Butler repo.
I believe this provides the same functionality and users can still choose the non-default masks by linking in their own reruns.