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
Given the CMIP6 closing deadline it is unlikely that CMIP6 will change. Having it r/w with a staging index will require a very large staging index for the collection. Instead it would be a greater benefit to consider adding the UK/EU/AU entries so they will remain discoverable by users of our sites. Just a thought if we are provisioning multiple indexes with very large capacities.
Perhaps we can use the CMIP6 staging initially and then decommission the index once the records are finalized? that way the resources could be re-provisioned.
The UK/EU/AU records can be retrieved from LLNL Solr as we have the replica shards from all.
The text was updated successfully, but these errors were encountered:
When is the closing deadline for CMIP6 publishing, and when we would transition it to read-only? Although we haven't formally discussed how we transition read-write to read-only. I think you are right, that transition would involve getting rid of the staging index used to publish and update metadata.
How many Dataset and File metadata entries are in the UK/EU/AU CMIP6? Would these metadata entries change and need to be refreshed once they were first loaded?
April 7 for "original" publication. April 20 for final LLNL replica publishing, other sites would sync (not a hard date but 90% likely to stick).
We would do the UK/EU/AU as a one-time load and that would be presumed read-only.
We will need to process retractions and data relocations as exceptional cases, which would apply to anything, as eg. any site could just drop out.
Given the CMIP6 closing deadline it is unlikely that CMIP6 will change. Having it r/w with a staging index will require a very large staging index for the collection. Instead it would be a greater benefit to consider adding the UK/EU/AU entries so they will remain discoverable by users of our sites. Just a thought if we are provisioning multiple indexes with very large capacities.
Perhaps we can use the CMIP6 staging initially and then decommission the index once the records are finalized? that way the resources could be re-provisioned.
The UK/EU/AU records can be retrieved from LLNL Solr as we have the replica shards from all.
The text was updated successfully, but these errors were encountered: