Uploaded image for project: 'Request For Comments'
  1. Request For Comments
  2. RFC-399

LDM-582: Lossy Compression WG Charge

    Details

    • Type: RFC
    • Status: Implemented
    • Resolution: Done
    • Component/s: DM, TCT
    • Labels:

      Description

      I request approval of the charge for the Lossy Compression WG.
      https://github.com/lsst/LDM-582

        Attachments

          Issue Links

            Activity

            Hide
            jbosch Jim Bosch added a comment -

            I'm by no means trying to stop acceptance of the charge, but a quick note, re "Acceptable Loss": given the way they average over things, I don't expect the SRD metrics to be nearly as sensitive to compression as shear tests (which Bob Armstrong, on the committee, of course knows all about).

            Show
            jbosch Jim Bosch added a comment - I'm by no means trying to stop acceptance of the charge, but a quick note, re "Acceptable Loss": given the way they average over things, I don't expect the SRD metrics to be nearly as sensitive to compression as shear tests (which Bob Armstrong , on the committee, of course knows all about).
            Hide
            price Paul Price added a comment -

            I am concerned that this working group has little allocated Project support: the membership of six consists of two observers and two people (myself and Bob Armstrong) who are not supported by LSST. While this is something I would like to contribute to (because we have the same concerns in HSC), my time cannot be guaranteed. I very much doubt we will be able to meet the very aggressive schedule set in the charge.

            Show
            price Paul Price added a comment - I am concerned that this working group has little allocated Project support: the membership of six consists of two observers and two people (myself and Bob Armstrong ) who are not supported by LSST. While this is something I would like to contribute to (because we have the same concerns in HSC), my time cannot be guaranteed. I very much doubt we will be able to meet the very aggressive schedule set in the charge.
            Hide
            xiuqin Xiuqin Wu [X] (Inactive) added a comment -

            Will the WG consider the lossless compression that was proposed in RFC-378 in terms of processing speed and storage size?

            Show
            xiuqin Xiuqin Wu [X] (Inactive) added a comment - Will the WG consider the lossless compression that was proposed in RFC-378 in terms of processing speed and storage size?
            Hide
            price Paul Price added a comment -

            The charge specifically relates to lossy compression, which will hopefully allow us to realise a compression factor of > 3 for our floating-point images, rather than the ~ 1.1 compression factor from lossless compression.

            Show
            price Paul Price added a comment - The charge specifically relates to lossy compression, which will hopefully allow us to realise a compression factor of > 3 for our floating-point images, rather than the ~ 1.1 compression factor from lossless compression.
            Hide
            mjuric Mario Juric added a comment -

            Paul Price, Robert Gruendl – I'll talk to John Swinbank to see what we can do about person-power. Good point, thanks for raising it (both of you).

            Show
            mjuric Mario Juric added a comment - Paul Price , Robert Gruendl – I'll talk to John Swinbank to see what we can do about person-power. Good point, thanks for raising it (both of you).
            Hide
            gruendl Robert Gruendl added a comment -

            With regard to loss-less compression. I see no reason not to obtain benchmarks but the anecdotal performance I know of when using GZIP_1, I think will preclude it from being a viable candidate at this time (the ultimate reason for this investigation has been to determine whether sufficient compression factors can be achieved to support storing and serving a broader range of reduced LSST image products (e.g. the PVI products) within the data releases.

            Show
            gruendl Robert Gruendl added a comment - With regard to loss-less compression. I see no reason not to obtain benchmarks but the anecdotal performance I know of when using GZIP_1, I think will preclude it from being a viable candidate at this time (the ultimate reason for this investigation has been to determine whether sufficient compression factors can be achieved to support storing and serving a broader range of reduced LSST image products (e.g. the PVI products) within the data releases.
            Hide
            womullan Wil O'Mullane added a comment -

            The last version in github master is good to go in DocuShare.

            Show
            womullan Wil O'Mullane added a comment - The last version in github master is good to go in DocuShare.
            Hide
            tjenness Tim Jenness added a comment -

            I updated the change record and released it on Docushare.

            Show
            tjenness Tim Jenness added a comment - I updated the change record and released it on Docushare.
            Hide
            gpdf Gregory Dubois-Felsmann added a comment -

            The report from the Working Group is available at DMTN-068.

            Show
            gpdf Gregory Dubois-Felsmann added a comment - The report from the Working Group is available at DMTN-068 .

              People

              • Assignee:
                gruendl Robert Gruendl
                Reporter:
                gruendl Robert Gruendl
                Watchers:
                Donald Petravick, Gregory Dubois-Felsmann, Jim Bosch, John Swinbank, Kian-Tat Lim, Krzysztof Findeisen, Mario Juric, Paul Price, Robert Gruendl, Tim Jenness, Wil O'Mullane, Xiuqin Wu [X] (Inactive)
              • Votes:
                0 Vote for this issue
                Watchers:
                12 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:
                  Planned End:

                  Summary Panel