Wednesday May 23 VConf.

From: JJ Kavelaars <jjk_at_hia-iha.nrc-cnrc.gc.ca>
Date: Wed, 23 May 2007 15:49:45 -0700

Here is the results of our brief discussion during todays VC.

There were a number of technical problems starting the meeting
and to keep to the time schedule we did not review old action items.

The next meeting is scheduled for June 6. However, JJK will be
away that week. Moving the Meeting to June 1st would be good.


Please read and provide corrections/additions.

All ACTIONS to be completed by the next VC unless otherwise
noted.


Data Transfer: Due to some software and hardware errors at the CADC
the transfer of stacks into the CADC archive has been stopped. CADC
is still pulling the files from Tpix into a local staging space
and leaving a copy on the TPIX server CLIX. Once the storage
problems at CADC have been resolved we will push the local
copy into the archive and remove the copy from CLIX.

ACTIONs:
        CADC will fix the archive problems and let TPix know
when the files currently on CLIX can be removed.

        Fred will continue to put new files into the e-transfer
area. These new files will be transfered to CADC and staged
until the storage issues are resolved at CADC.


CRC: CADC is now storing all TPIX stacks in uncompressed
form in the archive. This allows better interaction
and a substantial increase in performance when accessing
the stack collection inside VO tools that make 'cutouts' of
parts of the image data. The result of this change is that
CADC can not provide to Tpix a CRC for the compressed
stack [for TPIX to use to check that the stack we have is
the one they want us to have] since we are only storing the
CRC for the un-compressed image. For the intern TPIX will
need to uncompress the STACK at their end, compute the CRC
and compare the CRC of their uncompressed file to the CRC
that of the file in the CFHT storage system. The CADC should
have some mechanism for storing the CRC of the file that we
retrieved from TPIX [the compressed one] but currently we do not.

ACTION: None at this time.


LDAC: Stacks which do not have g/r/i filter images available will
not have a CHI2 image made and thus have no CHI2 catalog. For
the STACK releases we had intended to only provide 1 catalog per
stack/filter and this was to be the CHI2 catalog, limiting the number
of stacks for which catalogs are available. Providing a good
set of catalogs for the entire Wide coverage is highly desired and
so TPIX will send CADC single-filter catalogs for STACKs where
not CHI2 catalog exists. When users search/ retrieve the
stacks they will be given the option to get the 'single-filter'
catalogs from CADC when no CHI2 catalog exists. These single-filter
catalogs will be in LDAC format.

ACTIONs:
        Fred will place LDAC catalogs into the e-transfer stream.

        JJ will modify the interface so that T0004 LDAC catalogs are
        made available to users when CHI2 catalogs are not.

ldac2cat: Fred will provide a script that converts LDAC catalogs
to ASCII tables. This script will be provided to people who do
not want to handle the LDAC formated files and would rather have ASCII
tables to work with.

ACTION:
        Fed will provide script that can be distributed to LS users.

Help Pages: A text page that describes the content of the LS
stack products with links to information about how they are
created is needed.


ACTIONs:
        JJ will create help page and circulate.

        Yannick will provide an plot that summarizes the exposure and
        filter coverage of the Wide and Deep fields included in T0004.


Web Interface: A good / effective interface to the T0004 stacks
needs to be created. Using the T0003 page as a template is the
easiest solution for CADC. Is this page effective? Some users have
had difficultly using the interface to find the stack data they want.

ACTIONs:
        JJ will send the www interface link to the DOG and request that
        people provide feedback on improvements to the system.

        Yannick will provide some detailed feedback for the next TC.


Terapix Meta Data Tables; CADC will accept the CSV dumps the TP SQL
tables as the mode for transfer the TP metadata to CADC. Fred will
update the files periodically as new meta-data as the final T0004
stacks are produced.

ACTION:
        CADC will setup the metadata tables inside the CADC SYBASE system
        based on the tables provide by Fred.
Received on Wed May 23 2007 - 12:50:32 HST

This archive was generated by hypermail 2.3.0 : Thu Jul 27 2017 - 17:52:27 HST