[Halld-online] Requirements for the conditions database

[Elliott Wolin]

Dear Mark and Dmitry,

Below is an attempt to specify requirements for the conditions 
database.  Please get back to me if you need additional information...we 
can iterate as required.

Thanks,

Elliott


*


_Requirements for Hall D Conditions Database_*

Elliott Wolin
11-Nov-2011



_General Remarks_

The Hall D conditions database needs to store a wide variety of types of 
data relevant to configuring and running the apparatus and taking data.  
The information in the conditions database will be used by the offline 
reconstruction and analysis effort, and might further be used for 
debugging of online problems.  Further, it is important that this data 
be easily accessible from the analysis programs from both on- and 
off-site.  The data must be tagged appropriately, and the schema shown 
by Mark at the April 2011 Online meeting seems like it should work.

Note that the conditions database need not duplicate the function of the 
EPICS archiver, which records a vast quantity of fine-grained data from 
a large number of sensors.  Most of this data is not needed by the 
offline effort, but it is difficult to predict in advance which data 
will or will not be needed.  It does make sense, however, to record 
selected EPICS data in the conditions database that surely will be used 
by the offline, for example the solenoid setpoint current at the 
beginning of each run.


_Input Requirements_
_
_Data will be input to the conditions database in a variety of ways 
during experiment operations.  These include input from C++ and Java 
programs and from Python and shell (and perhaps Perl) scripts. Some 
input will come from long-running programs or scripts, some from 
short-lived programs or scripts launched as part of a run control 
transition, and some from CRON jobs.

Entry size will vary from single numbers or short strings to long 
strings containing many thousands of characters.

Entry frequency varies widely, with flurries of activity expected at the 
beginning and end of data taking runs, with relatively few entries 
during steady-state data taking.

Entry rates will peak within about 20 seconds after the start of a new 
run, with something like 10-20 processes making about 100-200 entries 
with a maximum aggregate size of less than 1 MB.  Note that how long 
these entries take is not critical as they will be made by programs and 
scripts running in the background.
_
_API requirements are not particularly strict, as only a few experienced 
members of the online group will be involved in making entries into the 
conditions database.

_

Retrieval Requirements_

These are largely driven by offline requirements, and are almost 
identical to those of the calibration database.  It is highly desirable 
that the API and access mechanisms for the conditions and calibration 
databases be as similar as possible.

An important consideration concerns export of the conditions database to 
other sites.  Some of the data entered will only be useful to the online 
group, although it is never clear whether any particular data type will 
be needed by the offline.  A mechanism needs to be provided that allows 
selective export of the database contents, e.g. by run range and data type.


-- 

				Sincerely,
					Elliott


================================================================================


  Those raised in a morally relative or neutral environment will hold
		    no truths to be self-evident.
				

Elliott Wolin
Staff Physicist, Jefferson Lab
12000 Jefferson Ave
Suite 8 MS 12A1
Newport News, VA 23606
757-269-7365

================================================================================


-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/halld-online/attachments/20111111/2bb48dd6/attachment.html