I am starting to think sticking with a rapid, monthly release
cycle would be a good thing. Here is how I see it for the
next few months (given month = end of the month)
* December:
- existing qserv prototype + 300-node test fixes +
Bill's tickets #3019, #3020
* January:
- new structure and new build system (Daniel)
- cleanup: (tbd who)
- #2528 (change header extensions)
- #2405 (centralize hardcoded constants)
- new packaging system (Fabrice)
- mysqldb wrapper (Jacek)
- #1933 (Ref*Match) (Serge)
- fine tune automated testing (Fabrice)
- query representation (Daniel)
* February:
- new logging (application level, and distributed) (Bill)
- #3087 (Central State System) (Jacek)
- data ingest (Serge)
- unit testing for various components
- new xrootd client integration (Daniel)
* March - June
- admin_client (Jacek)
- watcher/node installation (Jacek)
- data placement V1 (Andy+Jacek)
- worker scheduler
- cluster coordination (Jacek+Andy+Fabrice)
- secondary index (Daniel)
- result transfer protocol (Serge+Daniel)
- C++ geometry (Serge)
- Level 3 - key aspects - design (Jacek)
- more tbd
(plus, we need to deal with open tickets:
close/defer or assign to releases)
Let's discuss at the Qserv meeting tomorrow.
Jacek
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the QSERV-L list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1
|
