[Halld-online] Asynchronous scaler readout in Hall D

[Alexander Somov]

I think that we should follow the scheme which Hall-B guys are
using - they have much better experince than all of us as they
have spent quite time optimizing their DAQ performance.

Eventually, the final implementation scheme will depend on our
DAQ and trigger performance.

Having the data stream scalers is essential.The data stream scalers
should be accessable during data reprocessing and can be used by
the data quality monitoring tools at this stage (most experiment
have the special trigger types, like a minimum bias trigger to be
able to monitor occupancies during event reprocessing). Eventually,
I want to implement the special trigger type for them.

Cheers,
        Alex


On Tue, 7 Feb 2012, Elliott Wolin wrote:

> To all concerned with scalers in the Hall D Online,
>
> Hovanes, Dave and I met to consider many details of dealing with
> scalers. We propose the guidelines below. Note that these refer to
> normal data taking, not to code development/debugging efforts.  Further,
> the discussion below refers to asynchronous readout of scalers, not to
> how scalers are read out and inserted into the data stream.
>
>
> - the preferred method for reading scalers asynchronously (i.e. from
> registers) is to use a single process that writes to shared memory.
> This allows us to control resource usage on the ROC's and manage ROC
> processes.
>
> - this process is independent of the DAQ system. It should be the only
> processes that reads scalers from the registers, even the ROC code
> should not do this.
>
> - any other process that needs scaler data can get it from the shared
> memory. This includes an EPICS IOC process and perhaps other processes.
>
> - current thinking is to use an EPICS IOC to serve asynchronous scalers
> out to monitoring processes or to servers that bundle and then
> redistribute scaler and other data.
>
> - One question that arose was whether all crates should be part of the
> DAQ system. In particular, should the discriminator crates include a TI
> board, accept triggers, be part of the CODA DAQ configuration, etc?  In
> favor is the simplicity of putting discriminator scalers in the data
> stream and the fact that in this way it is possible to read scalers out
> of all crates at the same time.  The downside is that more crates will
> have to participate in the DAQ system and that there are other ways to
> get discriminator scalers into the data stream.
>
>
> Comments, questions, concerns...?
>
> Thanks,
>
> -- 
>
> 				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
>
> ================================================================================
>
> _______________________________________________
> Halld-online mailing list
> Halld-online at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-online
>